IJDIWC

IJDIWC
International Journal of
ISSN 2225-658X (Online)
DIGITAL INFORMATION AND WIRELESS COMMUNICATIONS
Volume 4 Issue 3
2014
TABLE OF CONTENTS
Original Articles
PAPER TITLE
AUTHORS
PAGES
CLOUD-LINKED AND CAMPUS-LINKED STUDENTS PERCEPTIONS Sivachandran Chandrasekaran,

OF COLLABORATIVE LEARNING AND DESIGN BASED LEARNING IN Guy
Littlefair,
Matthew
ENGINEERING
Joordens, Alex Stojcevski
267
AN APPLICATION OF DBR TO THE DEVELOPMENT OF A

Sei Sumi, Thomas Schalow
FRAMEWORK FOR ENGLISH LANGUAGE EDUCATION
276
AN ACTIVE MIDDLEWARE FOR SECURE AUTOMATIC S.

Kami
RECONFIGURATION OF APPLICATIONS FOR ANDROID DEVICES
Abdelrazek
284
Makki,
Karim
W.N.N.W. Marzudi, Z.Z. Abidin,

MINIMIZATION OF MUTUAL COUPLING USING NEUTRALIZATION
S.Z. Muji, Ma Yue, Raed A. AbdLINE TECHNIQUE FOR 2.4 GHZ WIRELESS APPLICATIONS
Alhameed
292
FRAMEWORK-BASED
MANAGEMENT
299
DATA
REQUIREMENTS
FOR
IT
TOP
LITERATURE-BASED AND ARCHITECTURE-BASED MODERNIZATION
Virpi Hotti
Virpi
Hotti,
Saarelainen
Miia-Maarit
305
A MEDIUM ACCESS PROTOCOL FOR COGNITIVE RADIO NETWORKS Dina Tarek Mohamed, Amira M.
BASED ON PACKET'S COLLISION AND CHANNELS' USAGE
Kotb, S.H.Ahmed
314
ICT GOVERNANCE: LITERATURE REVIEW AND SURVEY IN Virpi Hotti, Heikki Merilainen,
NORTHERN SAVONIA
Harri Salomaa, Antti Lahtela
333
PERSPECTIVE FOR SUGGESTION ABOUT COMMUNICATION PLAN /

Nattanun Siricharoen, Waralak
STRATEGY OF THAILAND USING FOR ASEAN ONE COMMUNITY
Siricharoen
RELEVANT TO THE 3 PILLARS: APSC, AEC, AND ASCC
341
AN INSTRUCTIONAL DESIGN MODEL AND CRITERIA FOR Mohamed Elsayed

DESIGNING AND DEVELOPING ONLINE VIRTUAL LABS
Shinobu Hasegawa
355
Ahmed,
APPLICATIONS OF THE ROTATING CALIPERS TO GEOMETRIC

Godfried Toussaint
PROBLEMS IN TWO AND THREE DIMENSIONS
372
Tiago Cinto, Harlei M. A. Leite,

VIRTUAL LEARNING ENVIRONMENTS: PROPOSALS FOR
Cecilia S. A. Peixoto, Dalton S.
AUTHORING AND VISUALIZATION OF EDUCATIONAL CONTENT
Arantes
387
MINIATURIZED
BANDSTOP
FILTERS
COMPLEMENTARY RESONATORS
401
USING
SLOTTED-
Mohammed Bait-Suwailam
International Journal of Digital Information and Wireless Communications (IJDIWC) 4(3): 267-275
The Society of Digital Information and Wireless Communications, 2014 (ISSN: 2225-658X)
Cloud-Linked and Campus-Linked Students Perceptions of

Collaborative Learning and Design Based Learning in Engineering
Sivachandran Chandrasekaran, Guy Littlefair, Matthew Joordens, Alex Stojcevski
Deakin University
75 Pigdons Rd, Waurn Ponds, Victoria 3216 Australia
schandra@deakin.edu.au, guy.littlefair@deakin.edu.au, matthew.joordens@deakin.edu.au,
alex.stojcevski@deakin.edu.au
ABSTRACT
The aim of this paper is to analyse and present cloudlink as well as campus-linked students perceptions of
collaborative learning and design based learning in
engineering. Project oriented design based learning
(PODBL) is a learning and teaching approach, where
students learn through design activities while being
driven by project(s). PODBL enhances cloud-linked
and campus-linked students ability to acquire career
essential skills that fulfill future industry needs. A
paper-based survey is used to recognise a cohort of
students' experience of collaborative learning and
design based learning in engineering. The paper-based
survey was given to 30 students from an engineering
discipline.
The quantitative analysis of the survey results shows
that more than 50% of the students view collaborative
learning to have a large benefit in design-based
learning.
KEYWORDS
Design based learning, project oriented design based
learning, collaborative learning, cloud-linked and
campus-linked.
1 INTRODUCTION
Collaborative learning is learning and teaching
approach that encourages students to work
together, share ideas, experiences and evaluate
each others results [1, 2]. In collaborative
learning, each member of the team needs to
understand what other team members have agreed
to research and what them as individuals must
contribute to the team. The students locate
resources that are directly relevant to the learning
issues. In project-oriented design based learning

approach, cloud-linked (off-campus) and campuslinked (on-campus) students work in teams of
four to six members with a facilitator. The same
group meets regularly throughout the trimester to
work on a series of design activities. The learning
and teaching delivery is a combination of cloudlinked and campus-linked learning activities.
Design based learning enables students to
evidence their achievement through collaborative
learning in cloud-linked and campus-linked study
mode.
Engineering at Deakin has used design-based
learning (DBL) as one of its engineering learning
principles for further development in teaching and
learning. Deakin University has about a third of its
students studying cloud linked [3]. It is required to
improve the learning and teaching process as a
holistic approach from the perspective of students
and staff over the entire degree program. The
qualitative and quantitative paper based survey
method is used to obtain cloud-linked and
campus-linked students perspectives. This
research paper shows a cohort of students'
perceptions of collaborative learning and design
based learning in engineering.
2 DESIGN BASED LEARNING
Design based learning (DBL) is a self-directed
approach in which students initiate learning by
designing creative and innovative practical
solutions which fulfil academic and industry
expectations. Design based learning is an effective
vehicle for learning that is centred on a design
problem solving structure adopted from a
267
combination of problem and project based
learning. Design projects have been used to
motivate and teach science in elementary, middle,
and high school classrooms and can help to open
doors to possible engineering careers. Design
based learning was implemented more than ten
years ago, however it is a concept that still needs
further development. With this in mind, it is very
important to characterise DBL as an educational
concept in higher engineering education [4] [5, 6].
A design based learning environment assists
curriculum to move into the twenty first century
with students being hands-on in their work, in
addition to using problem solving skills, engaging
in collaborative teamwork, creating innovative
designs, learning actively, and engaging with realworld assignments. Figure 1 illustrates the design
based learning process.
Figure 1. Design based learning process
By engaging students in learning design, DBL

provides an opportunity to experience individual,
inventive and creative projects that initiates the
learning process in relation to their preferences,
learning styles and various skills. Yaron Dopplet
[5] states that DBL is used to produce a
curriculum that improves learning for all students
in science education. Students are involved in
solving a problem through a creative project and
experience meaningful ideas that allows them to
analyse a suitable solution for it. To provide
students with better practise in design and
technology, DBL has several advantages that meet
social, economic and industry needs. It is also an
active learning process which makes students

practice and recognize different learning styles
and team based activity supports learning and
sharing through cooperative methods [6].
3 PROJECT ORIENTED DESIGN BASED
LEARNING (PODBL)
When students are involved in solving a problem
through a creative project, they will experience
meaningful ideas that allow them to analyse the
suitable solution for it. It is a basic quality for a
professional to deal with problems and find
solutions for these problems. Educational
institutions need to teach and train students not
only to be a problem solver, but also think about
achieving innovative and creative skills.
There are different kinds of problems exist in
engineering. One of these is that design problems
are most important to attract young and
imaginative students and projects are considered
to be the best way for students to interact with
teachers [7]. In addition to providing students with
better practice in design and technology, project
oriented design based learning will involve several
advantages such as when good design meets
social, economic and industrial needs [8]. This
active learning process makes students practice
and recognise different learning styles that support
learning and sharing through cooperative methods
[9].
The newly proposed approach, Project Oriented
Design Based Learning method is able to motivate
the students and teach engineering design in
classrooms to achieve more practical experience
that fulfill the academics and industry needs.
Project Oriented Design Based Learning is set to
have a positive effect on students knowledge of
the content and development of skills, such as
innovation and creativity that increases their
motivation and engagement [10-12].
In PODBL, learning is initiated by design based
learning through projects, which incorporates
aspects of design by active learning, learning by
doing, creative problem solving and innovative
designing [13]. Engineering programs are required
268
to demonstrate that their graduates are capable of
acquiring and achieving career focused learning
outcomes. It creates a boundary for a student
learning capability, when programs are content
driven that is focused on engineering science and
technology courses. PODBL is a structured
framework, which will overcome insufficiency of
design practice related to industry requirement.
3.1 Cloud-Linked and Campus-Linked
The cloud-linked (online education) has played an
important role in the provision of educational
equity for learners who live in remote Australian
communities [14]. Engaging students and lectures
will always ensure a positive experience in offcampus education (cloud-linked) [15]. The
teaching staff team must be committed, equipped
and
adequately
resourced
to
support
implementation of the cloud-linked (distance
based learning) mode.
Cloud-linked must enhance learner-staff and
learner-learner interactions as well as enriching
the learning experiences of both learners and staff
[16]. The technology will never replace the
teacher in any classroom, but with careful
planning and judicious use. It can provide ways of
enhancing rural education. The Problems exists in
distance education are quality of instruction,
misuse of technology, attitudes of instructors,
students and administrators. To fulfill cloud-linked
(distance learning) students expectations, the
Internet and compressed video have taken distance
learning in new directions, allowing distance
learning to occur in real time [17].
Engineers Australia, Accreditation of engineering
programs in Australia recognizes that educational
and learning techniques are continually changing
and advancing [18]. Engineers Australia policy on
Accreditation of programs offered in distance
mode guidelines states, Electronic and face-toface opportunities must be provided for distance
mode learners to interact, particularly to ensure
that group and team based learning experiences
are equitable for both campus-linked (on-campus)
and cloud-linked (online mode) cohorts and
The learning and assessment design, learning

activities, learning resources and assessment
measures for a distance mode implementation of a
program must be purpose built to support the
external learner in a comprehensive and
independent manner [16]. The key to successful
learning and teaching is to create flexible systems
(a sustainable online content development model)
that can accommodate changes in content,
technology, and student needs [19].
3.2 The PODBL Cycle
The PODBL model is applied across all four years
of engineering and across the four disciplines of
civil, mechanical, electrical and mechatronics
taught in the school of engineering, Deakin
University. Students are introduced to projects
from the first year of engineering; the projects in
the first year are university projects and as the
students progress in year two through to year four
projects from the industry are introduced.
In the PODBL model participants work in teams
four to six members with a facilitator, which is
similar to the project based learning approach. The
same group meets regularly throughout the
trimester to work on a series of design activities.
Figure 2. The PODBL cycle
269
The PODBL cycle (figure 2) involves nine main
steps. The steps are illustrated in figure 2 above.
Steps 1-6 & step 9 are a combination of both cloud
and located learning (campus-linked), and steps 78 are performed through located learning. Steps 1
to 3 of the PODBL cycle the project is presented
to the students by the academic or industry based
facilitator and the students are given the
opportunity to brainstorm on the project to
identify the problems and engage in concept
research to understand the essential learning issues
and the overlapping issues. These activities are
conducted in the seamless digital environment for
cloud-linked learning. The students are provided
with integrated short, accessible, highly visual,
media-rich, interactive learning experiences
rebuilt for the mobile screen, and integrating
learning resources created by Deakin and other
worldly universities and premium providers.
The school of engineering has already started in
this direction with the use of technology to enable
learning practice and under this initiative
encouraged staff to record the lectures and
tutorials. The staff could record their lecture on
video or use screen capture software to record the
lecture slides along with audio and the recording
are provided to the students via the cloud. These
lecture resources provide the students with an
opportunity to revisit lectures and go through the
concepts discussed during the session. The lecture
videos allow the students to catch up on missed
lectures with an experience similar to attending a
lecture. Students have indicated in a survey they
use the lecture as a catch up exercise and it allows
them greater interaction [20]. The school is
moving towards the recording of short topic based
clips no longer than ten minutes which will allow
the students to access material which are media
rich and visually engaging.
Steps 4 to 6 in the PODBL cycle the teams are
involved in developing the design brief to produce
the key planning document with the specifications
for the project and project plan, select the concept
for the solution based on the ideas generated
during the concept research step and move on to
the design and development step to develop the
final design. These activities are a combination of
cloud and located learning activities. The students

will interact with the staff during the design brief
and concept selection stage; this interaction can be
over the cloud or on campus for students who are
able to come on campus.
The school of engineering has taken a step in this
through the use of online tutorials. Tutorials
conducted in the classroom provide the students to
interact with the staff member and also revisit
concepts covered in the lectures and collaborate
with staff. This initiative of collaboration and
interaction between the students and staff has been
taken online through the use of Elluminate Live!
eLive, a technology resource which facilitates
communication and collaboration between staff
and students. It allows the staff and students to
talk over the Internet and also via an online chat
room and to have online meetings and facilitates
learning and training. It presents cloud-linked
students to interact and collaborate with the staff
and their peers in a safe and secure environment.
Staff members can share audio and visual
materials with the participants and can also invite
guest speakers like experts from the industry. The
flexibility of the online environment allows for the
meeting to set up without the boundaries of time
and space. The initiative from the school has been
well received by the cloud-linked students and
also by the campus-linked students who use this as
an extra opportunity to collaborate with the staff
and their peers.
Project oriented design based learning approach
focuses on this interactivity between the staff and
students and among themselves and this resource
provides them opportunity to interact in various
setting in which members from the industry can
also be invited to share their ideas and views. This
resource also allows the school to provide the
enhanced interaction between student and staff as
mentioned in the cloud learning policy.
The project in the PODBL model allows for the
learning and teaching delivery to take place as a
combination of cloud-linked and campus-linked
learning activities. Cloud learning enables students
to evidence their achievement and requires
students to be generators of content, collaborators
270
in solving real world problems, and evidence their
achievements in professional and personal digital
portfolios. With premium cloud learning
experiences in place, students who come to
campus will have the opportunity to engage with
teaching staff and peers in opportunities for rich
interpersonal interaction through large and small
team activities and also provided through the use
of lecture videos, online tutorial for students who
are able to make it campus-linked.
4 COLLABORATIVE LEARNING
When students work in groups of two or more
where the centre of attention is project report or a
design is known as collaborative learning.
Participants have their individual accountability
along with several conditions. Collaborative
learning requires working together towards a
common goal where students are responsible for
one anothers learning [21]. It is an educational
approach to learning and teaching that involves a
group of learners working together to solve a
problem, complete a task or create a product [22].
With different learning styles students are able to
express their skills and talents through working on
projects or by simply designing experiments in
authentic learning environments [4, 23].
Integrating design and technology tools into
science education provides students with dynamic
learning opportunities to actively investigate and
construct innovative design solutions. By
engaging students in collaborative learning,
PODBL provides an opportunity to experience
individual, inventive and creative projects that
initiate the learning process in relation to their
preferences, learning styles and various skills.
4 METHODOLOGY
Engineering at Deakin has used design-based
learning (DBL) as one of its engineering learning
principles for further development in teaching and
learning. It is required to improve the learning and
teaching process as a holistic approach from the
perspective of students and staff over the entire
degree program. The qualitative and quantitative
paper based survey method is used to obtain
cloud-linked and campus-linked students

perspectives. Qualitative methods are useful for
evaluating, developing program goals and for
involving participants in the evaluation process to
gain their insight and perspective [24].
From the quantitative and qualitative analysis
performed, the results are analysed and presented
from a students perspective about collaborative
learning and design based learning in engineering
curriculum. The survey is paper based which was
conducted by a third person not involved in the
research project. The survey was given to cloudlinked/campus-linked students in the third year of
engineering and was anonymous and nonidentifiable. These results are from students own
experiences and the results present various views,
which include students knowledge and
expectations of collaborative learning and design
based learning in engineering.
5 RESULTS
5.1 Students Perceptions of Collaborative
Learning
The cloud-linked and campus-linked students
views of collaborative learning through project
oriented design based learning in this research
come from a cohort of senior year undergraduate
engineering. This study goal is to determine the
students perspective of DBL and collaborative
learning and how the perspective changes over the
years studying engineering.
Table 1 shows students' perceptions of advantages
of teamwork (collaborative learning) in designbased learning, which includes real world
experience, teamwork and interaction. This
indicates that the present curriculum needs a
change in teaching by implementing the DBL
units from senior year engineering programs.
About 30% of campus-linked and 4% of cloudlinked students says through teamwork DBL, they
acquired interactive knowledge, 15% of students
mentioned that it develops collaborative skill,
management skill and social science. Its
interesting to see 11% senior year students says
that they get the opportunity of managing large
271
projects through real world problems with
industrial experiences.
Table 1: Advantages of collaborative learning
Cloud
linked
(%)
4
Campus
linked
(%)
11
Good experience, Time efficient
The work is divided, faster to

give results
Team work, interactive
knowledge
Develops Collaborative,
management skills, social skills
Real world experience, like
industry experience, build large
projects
30
15
Figure 3: Modes of Design based learning preferred
No answer
The quantitative analysis of the survey results

shows about more than 50% of the students view
some aspect of the advantage of teamwork
(collaborative learning) in design-based learning
includes real world experience and interaction,
develops collaborative, management and social
skills. Overall students views resembles that most
of the essential graduate abilities are attained
through collaborative learning (teamwork) in DBL
mode. The students perspectives are required to
verify the learning and teaching methods and to
identify the best practices in these methods to
ensure the best possible learning experiences for
the student.
Students perceptions
More discussion Opportunities,
distribution of knowledge
Table 2 illustrates students perceptions on

disadvantages of (teamwork) collaborative
learning through design-based learning. Most of
the cloud-linked and campus-linked students
mentioned the problems exist in teamwork such as
lack of consistence, communication, decisionmaking, co-operation and time management.
When students are questioned about their
preference of learning design based education
through collaborative learning. Figure 3 clearly
shows 52% of campus-linked and 9% of cloudlinked students preferred half teams and half
individuals mode of design based learning
through collaborative learning.
Table 2: Disadvantages of collaborative learning
Cloud
linked
(%)
0
Campus
linked
(%)
27
10
12
No individual effort, loss of time
15
Hard to make decisions and

organize
Lack of consistence, slack
students
16
10
Team members are not
contributing
Independent learning not
applicable, Relaying on other
tasks
5.2 Students Perceptions on Design Based

Learning
The students views on design-based learning
(DBL) in this research come from cloud-linked
and campus-linked students in undergraduate
engineering. The study goal is to determine the
students perspective of DBL and how the
perspective changes over the cloud-linked and
campus-linked students studying engineering. The
way engineering students tackle their university
degrees are somewhat very different to the way
engineering students went through their studies
few years ago. The tables and figures below show
a comparison of cloud-linked and campus-linked
students perspectives.
272
Table 3: Design based learning means
Campus
linked
(%)
22
19
31
10
Cloud
linked
(%)
No answer
Learning in a particular way,
Learning by doing, hands on
projects
Learning through Practical
application of engineering (design
problem)
Project based learning (real world
projects), learning through
projects (design projects)
Figure 4: DBL final year project
When students are asked about the influence of

DBL in their future career and importance of DBL
in their final year project. Figure 4 show that
(39%) campus-linked students and (10%) cloudlinked students perceived that DBL is important in
their final year project. Almost 22% of campuslinked and 5% of cloud-linked students strongly
maintains that DBL is necessary in their final year
project.
Table 3 shows students perception of design based

learning approach, which is used as current
curriculum methodology at Deakin engineering.
Approximately 3% to 19% of students in both
study mode revealed that DBL is learning in a
particular manner, learning by doing, hands on
project. About 18% of cloud-linked and campuslinked students says that DBL is a project based
learning (real world projects), learning through
projects (design projects). The cloud-linked and
campus-linked students' perception on DBL gives
an encouraging sign for the engineering
curriculum educators in the School of Engineering
at Deakin University.
Table 4: DBL in the curriculum
No answer
Figure 5: DBL to your career
It is clearly shown in figure 5, campus-linked

(40%)
and
cloud-linked
(8%)
students
recommended DBL is necessary for their future
engineering career. Almost 25% of campus-linked
and 7% of cloud-linked students strongly
maintains that DBL is very important in their
future career.
Practical Learning, Demos in

classes, more practical less
theory
Learning through projects,
design techniques
Integrate with teaching
(Individual DBL units)
Labs, practicals, lectures,
tutorials, more assessments
based on practicals
Cloud
linked
(%)
0
Campus
linked
(%)
37
19
10
11
Table 4 indicates cloud-linked/campus-linked

students' perceptions of DBL in their curriculum,
almost 19% of students from campus linked and
7% from cloud-linked experienced DBL as
practical learning, demos in their classroom.
273
About 7% of all students in both study mode
believed DBL through industry related projects;
learning through projects, design techniques.
Through a design based learning approach, design
is driven by qualitative thinking, speculation,
ideation, prototyping and specification [25, 26].
To achieve the goal of developing education,
design educators provide practical strategies that
exhibit the pedagogy of design education and
problem solving processes. The goal of the design
approach was not to change the whole curriculum
of art education but for students to develop the
ability to enhance and transform ideas through
visualization, manipulation and the application of
data to problem solving through design activities
[27, 28].
The purpose of design education is to enhance
learning in order to teach students to become
active participants to solve the design problems
around them [29, 30]. The aim of all educational
institutions is to educate students as graduate
engineers, and equip them with the ability to work
in industry with graduate ready skills such as
creativity, innovation, teamwork, problem solving,
observation, analytical thinking, communication,
and prototyping etc. Every learning and teaching
approach has its own special way of confronting
engineering problems. In design based learning,
students solve engineering problems by using
design as a vehicle.
6 CONCLUSION
This research paper focused on identifying cloudlinked and campus-linked students perceptions on
collaborative learning and design-based learning
in engineering. The quantitative analysis of the
survey results shows that more than 50% of the
students view collaborative learning to have a
large benefit in design-based learning. Project
oriented design based learning approach created
an enhancing environment for students and staff
members through collaborative learning in cloudlinked and campus-linked study mode. The
students' perceptions always have an important
value in an engineering curriculum to foster their
learning outcomes.
7 REFERENCES
1.
Laal, M., and Laal, M.: Collaborative learning: what is

it? Procedia - Social and Behavioral Sciences, vol.31,
pp. 491-495. Elsevier (2012).
2. Laal, M., Laal, M., and Kermanshahi, Z.K.: 21st
Century Learning; Learning in Collaboration. Procedia
- Social and Behavioral Sciences, vol.47, pp. 16961701. Elsevier (2012).
3. Palmer, S., Sharyn, B.: Academic performance and
persistence of on- and offcampus engineering and
technology students. in Engineering education for a
sustainble future: proceedings of the 14th Annual
conference for Australasian Association for
Engineering Education and 9th Australasian Women in
Engineering Forum. Melbourne (2003).
4. Wijnen, W.H.F.W.: Towards Design-Based Learning,
in Educational Service Centre, Technische Universiteit
Eindhoven (1999).
5. Doppelt, Y., Assessing creative thinking in designbased learning. International Journal of Technology
and Design Education, 19(1): p. 55-65 (2009).
6. Dopplet, Y., M.M.M., Schunn, C.D., Eli Silk, and
Denis Krysinski, Engagement and Achievements: A
case study of Design-based learning in a science
context (2008).
7. Chandrasekaran, S., Stojcevski, A., Littlefair, G.,
Joordens, M.: Learning through Projects in Engineering
Education. in 40th SEFI Annual Conference
Thessaloniki, Greece (2012).
Joordens, M.: Design Based Learning - Students Views
on Industry Requirements, in International Symposium
on
Project
Approaches
in
Engineering
Education(PAEE),
Eindhoven
University
of
Technology, the Netherlands (2013).
Joordens, M.: Project-Oriented Design Based Learning:
Aligning Students Views With Industry Needs,
International Journal of Engineering Education, vol.29,
pp. 1109-1118, Tempus, Great Britian (2013).
10. Joordens, M., Chandrasekaran, S., Stojcevski, A.,
Littlefair, G., The Process of Design Based Learning:
A Students' Prespectives. in Australasian Association
for Engineering Education (AAEE)
Annual
Conference, Melbourne (2012).
Joordens, M.: Project Oriented Design Based Learning
- Staff Perspectives. in The 4th International Research
Symposium on Problem-Based Learning (IRSPBL),
Malaysia (2013).
Joordens, M.: A Comparative study of students
perceptions on project oriented design based learning
in Engineering education. in 2013 Australasian
Association of Engineering Education conference,
Goldcoast (2013).
274
Joordens, M.: Accreditation Inspired Project Oriented
Design Based Learning curriculum for Engineering
Education, in 2nd International Engineering and
Technology Education Conference (IETEC), Ho Chi
Minh City, Vietnam (2013).
14. Stevens, K., Australian Developments in Distance
Education and Their Implications for Rural Schools.
Journal of Research in Rural Educaion, vol.10, pp. 7883 (1994).
15. Rachel Crease, B.P.a.L.H. Bridging the gap - engaging
distance education students in a virtual world, in
Proceedings ascilite 2011 Changing Demands,
Changing directions, Hobart (2011).
16. Bradley, P.A., Engineers Australia Policy on
Accreditation of Programs Offered in Distance Mode,
Engineers Australia, Melbourne (2011).
17. Valentine, D., Distance Learning: Promises, Problems,
and Possibilities. Online Journal of Distance learning
Administration, vol.5, (2002).
18. EA, Stage1 competency standard for professional
engineer, Engineers Australia: Australia (2012).
19. Stuart Palmer, W.T., The evolution of online teaching
and learning in engineering at Deakin University,
Journal of Computing in Higher Education, vol.3, pp.
91-109 (2001).
20. Joordens, M., Chandran, J., Stojcevski, A.: Comparison
of Technology Enabled Learning Practices (TELP) in
Engineering: a student's perspective, in 23rd Annual
Conference of the Australasian Association for
Engineering Education. Melbourne (2012).
21. Dooly, M.: Telecollaborative Language Learning. A
guidebook to moderating intercultural collaboration
online, in Constructive Knowledge Together, Bern,
Editor. Peter Lang (2008).
22. Laal, M., Ghodsi, S.M.: Benefits of collaborative
learning. Procedia - Social and Behavioral Sciences,
vol.31, pp. 486-490. Elservier (2012).
23. Doppelt, Y., Schunn, C.D.: Identifying students

perceptions of the important classroom features
affecting learning aspects of a design-based learning
environment, Learning Research and Development
Center (LRDC), University of Pittsburgh (2007).
24. Hammel J, Royeen, C.B., Bagatell, N., Chandler, B.,
Jensen, G., Loveland, J., Stone, G.: Student Prespective
on Problem-Based Learning in an Occupational
Therapy Curriculum: A Multiyear Qualitative
Evaluation. American Journal of Occupational
Therapy, vol.53, pp.199-206, (1999).
25. Perrenet, J., Aerts, A., Van der Woude, J.: Design
Based Learning in the Curriculum of Computing
Science - a Skillful Struggle, in Proceedings of 2003
International Conference on Engineering Education
(2003).
26. Dym, C.L.: Design, systems, and engineering
education. International Journal of Engineering
Education, vol.20, pp. 305-312 (2004).
27. Deakin, Deakin Design Forum : Industry and
Academia needs. Deakin University, Australia (2012).
28. Litzinger, T., Lattuca, L.R., Hadgraft, R., Newstetter,
W.: Engineering Education and the Development of
Expertise. Journal of Engineering Education, vol.1,
pp.123-150 (2011).
29. Lehmann, M., Christensen, P., Du, X., Thrane, M.:
Problem-oriented and project-based learning (POPBL)
as an innovative learning strategy for sustainble
development in engineering education. European
Journal of Engineering Education, vol.33, pp 283-295
(2008).
30. Atman, C.J., et al., Engineering design processes: A
comparison of students and expert practitioners.
Journal of Engineering Education, vol.96, pp. 359-379
(2007).
275
An Application of DBR to the Development of

a Framework for English Language Education
Sei Sumi and Thomas Schalow
University of Marketing and Distribution Sciences
Gakuen Nishimachi 3-1, Nishi-ku, Kobe-city, Hyogo 651-2188 Japan
seisumi@lang-tech.net ijinkan@mac.com
ABSTRACT
This paper explains how to develop an integrated
framework for English language education based on
the four phases of design-based research (DBR). In the
first phase of the DBR we conducted a qualitative
investigation to identify relevant problems at the
Japanese university where this research was
undertaken. To ameliorate these problems in the
second phase, we (a) developed an original placement
test, (b) created a classification technique by
employing a Bayesian algorithm, and (c) established an
e-learning system and comprehensive educational
policy. In the third phase the developed placement test
was tested, and all the test items were analyzed using a
Rash model. In the last phase our framework for
English language education was put into practice. Our
findings confirmed that the new framework not only
resulted in a substantial reduction in budgetary and
manpower costs, but also successfully contributed to
the improvement of the students learning outcomes
and integration of the entire process of English
language education.
KEYWORDS
English language education, design-based research, test
development, e-learning, education in Japanese
1 INTRODUCTION
English language education in Japanese
universities has been changing rapidly in recent
years as a result of a number of government
initiatives and directives. The Ministry of
Education, Culture, Sports, Science and
Technology recently released a report on the
globalization of Japanese universities.[1] In the
report reforms in English language education were
particularly emphasized. These reforms were in
line with Japanese government plans to employ

standardized tests such as TOEFL to set minimum
English requirements for new candidates for
positions as government bureaucrats.[2] These
changes will have a tremendous impact on the
shape of English language education at Japanese
universities.
In addition to changes resulting from these
government initiatives, change is being driven by
the sharp decline in the student population.
Japanese universities have started lowering their
standards of admission and the level of exams to
enroll as many students as possible. There are
even many students who never take any kind of
test to get into universities, if they possess a talent
for sports or music. Consequently, it has become
difficult to require students to have a basic
command of English. Some students have not
attained even a high school level of English ability.
Many universities are addressing this challenge by
offering small-group lessons and remedial
education classes based on students English
ability. A key common concern for these
universities is how to raise fundamental English
ability in areas such as vocabulary and grammar.
To respond to these changes in the direction of
English language education our university, the
University of Marketing and Distribution Sciences,
has initiated several projects.[3] In 2013 we
developed and deployed a novel framework for
English language education based on a
comprehensive assessment of English language
education, from the placement test for freshmen to
the final achievement test for sophomores. This
paper reports on our project.
276
2 DESIGN-BASED RESEARCH
The framework for our English language program
was developed using design-based research (DBR).
The origin of DBR is most closely associated with
Brown (1992) and Collins (1992).[4] [5] In the early
stages of development DBR was labeled a design
experiment (DE) in consideration of the research
focus, practice, and underlying epistemology.[6]
DE focuses on the understanding of a learning
ecology, which consists of complex elements of
an interactive system in teaching and learning,
with different types and levels.[7] In contrast to
other research methodologies that reduce complex
teaching and learning practices to a simple causeand-effect models with a list of separating factors,
DE aims to offer a holistic explanation for designbased practice, and has greatly influenced
educational studies.
DE has grown in diversity since it first appeared in
the field of educational studies. Consequently, it
has become difficult to find theoretical or
methodological
coherence
among
efforts
purporting to use DE. In recognition of this
diverse nature of DE, the Design-Based Research
Collective (2003) was created to provide new
theoretical and methodological advances.[8] Their
work clearly defined the character of DBR, and
showed the direction for ensuing studies based on
DBR.
DBR, with its flexible and practical characteristics

(Cobb, Confrey, diSessa, Lehrer, & Schauble,
2003), has proven especially useful in educational
and other settings that face new challenges
brought on by technological innovation.[9] DBR is
therefore now practically applied to teaching
practices, and though the number of such teaching
practices is quite limited (Hung, 2011), their
variety is growing.[10] For example, in the field of
mathematics education Gravermeijek and Cobb
(2006) applied a DBR framework to teaching
practice and contributed to identifying a local
instruction theory.[11] Yutdhana (2005, p. 170)
introduced DBR as an effective alternative to
Computer-assisted Language Learning research
and described its two key features as follows[12];
1. Designing learning environments and
2. Developing theories of learning.
According to Amiel and Reeves (2008), DBR
comprises four phases of study (Figure 1).[13]
Following these four phases, we conducted a
qualitative research study, designed a solution,
tested it, documented the results, and provided our
assessment of their theoretical and practical
implications.
2.1 Phase 1: Analysis of practical problems
In Phase 1 we conducted interviews to identify
Figure 1. The four phases of DBR (adopted from Amiel & Reeves, 2008, p. 34)
277
problems in the placement test the university had

used during the 2012 academic year. The
university had used a private, standardized test
every year since 2008 to divide freshmen into
three levels based on their scores; advanced
classes (20%), intermediate classes (60%), and
basic classes (20%). After the placement test
students take compulsory English classes at this
level for two years.
Many Japanese universities presently use private,
standardized tests such as TOEIC or TOEFL as
the basis for their placement tests.[14] These tests
are well known among Japanese students and
private companies in Japan, and have many
advantages. Perhaps the main advantage for
universities is that they offer a standardized, if not
always accurate, means to assess the language
abilities of students. Students were thought to
benefit by being placed in language classes they
seem to be best suited for. For these reasons, we
used a standardized test for the placement test, as
well as for the achievement test. The achievement
test is conducted two years after the placement test,
after students finish their compulsory English
lessons.
However, student interviews revealed a number of
problems with the placement test. Since the
English language abilities of students vary widely,
many students felt burdened with the length of the
exam, the unfamiliar testing style, and the highlevel test items. The university also felt
encumbered by the need to use a standardized
placement test because the month during which
the placement test is conducted coincides with the
busiest time for enrolling new students. The
university expends considerable manpower and a
substantial amount of financial resource to
conduct the placement test. The placement
examination results are returned a couple of days
after the test is conducted, and manpower resource
must once again be expended in checking results
and dividing all the students into the three levels.
Moreover, in addition to the fact the placement
test imposed a burden on the students and
administrative staff, all test items remain
undisclosed to the public, preventing the
university from identifying students weaknesses
and using the data for educational purposes after

the exam.
The original purpose of the placement test was
simply to divide students into three levels based
on their scores. To achieve this purpose the
students and administrative staff have thus far
been willing to accept the burden of the
examination. Through the interview process,
however, we found that students are questioning
the validity of using a standardized test for the
placement examination, while the university is
questioning the benefits of dividing the students
into skill levels that may not reflect working
knowledge of the language. Taking all these
problems into consideration, we decided to
develop an original placement test and devise a
framework for English language education that is
compatible with the diverse English language
abilities of students, thereby contributing to an
improvement in the students learning outcomes.
2.2 Phase 2: Development of solutions
In Phase 2, in an effort to ameliorate the problems
identified in Phase 1, we (a) developed an original
placement test, (b) devised a classification
technique employing a Bayesian algorithm, and
(c) established an e-learning system and language
education policy to take student needs and abilities
into consideration.
a. Development of the placement test
In order to develop an original placement test, we
decided to put the entrance examinations used by
our university from 2004 to 2012 into digital
format, creating a databank of approximately
1,000 test items. All test items were arranged
according to the correct answer rate, and 10% of
the total items were randomly selected. We called
this test the test.
We recognized numerous advantages of recycling
past entrance examination test items for the
placement test. All the test items for the entrance
examination represent the universitys minimum
standard of English. The vocabulary and grammar
employed in each test item also adhere to a
278
minimum standard. By using similar entrance

examinations every year, we are able to measure
students English ability over time. Moreover,
because all entrance examinations are published
and available to the public in the following year,
we are able to use the results to better understand
the students weaknesses at both the individual
and class levels.
b. Development of the classification technique
Several methods exist for classifying students into
different groups based on test scores. The simplest
approach is to use the raw score. However, using
this method it is impossible to discern a
meaningful difference between identical scores. If,
for example, students A and B each completed 50
of 100 questions on a test, but student A chose to
answer more difficult test items, while student B
was only able to answer much easier test items,
they might both achieve similar scores when the
examinations were graded. It would probably be
wrong to conclude, as raw test scores would, that
the English ability of student A and B is the same.
To avoid this problem, we applied a naive Bayes
classifier to the categorization of the students test
scores. A naive Bayes classifier allows users to
classify data based on a probability model that has
been widely used in solving problems such as the
classification of spam email. With this algorithm
we can classify students based on their test answer
patterns.
A classification technique that enables us to
categorize students into three levels can be
achieved by a developing Bayes theorem (1) for
an advanced class (2), an intermediate class (3),
and a basic class (4).
P(Hi|Dj) = P(Dj|Hi)P(Hi) / P(Dj)
(1)
P(Dj|H1)P(H1) > P(Dj|H2)P(H2) > P(Dj|H3)P(H3)
(2)
P(Dj|H1)P(H1) < P(Dj|H2)P(H2) > P(Dj|H3)P(H3)
(3)
P(Dj|H1)P(H1) < P(Dj|H2)P(H2) < P(Dj|H3)P(H3)
(4)
In the equations, P(H1), P(H2), and P(H3) are prior

probabilities. P(H1) is for advanced classes, P(H2)
is for intermediate classes, and P(H3) is for basic
classes. A prior probability indicates a correct
answer rate at each level on the placement test. A
prior probability can be subjectively set after
consideration of factors such as the students
scores at each level, instructors expectations, and
test difficulty. An advantage of Bayes theorem is
that it enables users to employ subjective factors
in order to build a probability model. The P(Hi|Dj)
section in the equations indicates the likelihood of
each test item, which informs us of level i,
indicating the students correct answer rate on a
test item j.
A probability model comprising prior probability
and likelihood can be trained through practical
application using Bayesian updating. The trained
probability model is applicable to a number of
fields, such as the development of a curriculum
and materials. The theoretical background of
Bayes theorem, allowing us to employ users
expectations and subjectivity for updating the
model, has many similarities with the DBR
framework, which integrates research and practice
to draw a better solution by following four phases.
c. Developing an
instructional design
e-learning
system
and
The e-learning system we developed also used the

databank of all past entrance examination test
items, producing enough lessons for two years of
compulsory study. To achieve this, we rearranged
all test items by the correct answer rate, divided
them by 20 test items, and successively created 48
sets of test items. In addition, we divided the 48
sets of test items into four categories, and named
each category Basic, Standard, Intermediate, and
Advanced. The following list indicates the four
categories and sets of test items included. The
percentage in the brackets shows the range of the
correct answer rate.
Basic:
Standard:
Intermediate:
Advanced:
SET001 to SET012 (99.1%66.4%)

SET013 to SET024 (66.3%52.7%)
SET025 to SET036 (52.6%39.5%)
SET037 to SET048 (39.4%3.2%)
279
During their two years of compulsory English

lessons students study English I in the first
semester of the first year, English II in the second
semester of the first year, English III in the first
semester of the second year, and English IV in the
second semester of the second year. We allocated
the test items of Basic to English I, Standard to
English II, Intermediate to English III, and
Advanced to English IV. Students spent
approximately 20 to 30 minutes studying each set
of test items every week. The test items studied in
the prior week were tested the following week.
The instruction design was designed based on the
Cyclic Model of Learning.[15]
The cyclic model of learning (CML, Takeuchi,
2007)[16] was designed with a special focus on the
local context, namely, Japanese EFL classrooms.
The most distinctive feature of the CML is that it
integrates in-class teaching practices with
students out-of-class self-learning, with the aid of
technology. In addition, the CML attempts to elicit
student participation in lessons and to promote
self-learning outside the classroom, both of which
are considered to be indispensable to successful
EFL learning (Takeuchi, 2002).[17]
All test items were uploaded onto a Moodle web
system.[18] Figure 2 shows an example of the Basic
test items. In addition to the test items, videos of
study-abroad programs, and learning materials for
basic grammar, vocabulary and extensive reading
were uploaded onto the Moodle website. Any
computer or smart phone with online capability
was able to access the study and test contents.
A final examination was conducted at the end of
each semester. Students thus take achievement
tests five times, including the placement test,
during their two years of compulsory English
lessons. All test scores are analyzed according to
levels, classes, faculties, and tutors every time the
final examination is conducted. The test results are
then applied to the improvement of test items,
materials, and lessons.
Figure 2. Example of Basic test items
2.3 Phase 3: Testing and refinement of solutions

In Phase 3, the test was tested, and each test
item was analyzed (N = 945, M = 41.19, SD =
9.31). Figure 3 shows the overall picture of the
result. All test items were analyzed using a Rash
model. The findings showed the average of the
examinees ability was -0.47, the standard
deviation was 0.54, and the reliability of the
examinees ability was .78. The average item
difficulty level was 0, the standard deviation was
1.32, and the reliability of the item difficulty level
was .97.
Figure 4 displays a persons and item map, which
allows us to understand the result of the test on a
single scale. As shown in the figure, misfit items
can be identified both in the area of the high- and
low-difficulty levels. While misfit items found at
the low-difficulty level had to be refined, misfit
items found at the high-difficulty level were items
that all students need to study throughout the twoyear compulsory English lessons. Therefore, we
retained those items without modification. No
misfit item (infit MSNQ 0.75) was found. Based
on the results, we decided to employ 100 modified
test items, and developed the placement test (ver.
1.0).
280
3 PRACTICE
Figure 3. Result of the test.
Figure 4. Person and item map
In Phase 4, the placement test (ver. 1.0; N = 653,

M = 37.31, SD = 11.45) was conducted on
freshmen entering in the academic year that began
in 2013 (March, 2013). According to an analysis
of a Rash model, the average of the examinees
ability was -0.62, the standard deviation was 0.60,
and the reliability of the examinees ability
was .84. In addition, the average item difficulty
level was 0, the standard deviation was 0.90, and
the reliability of the item difficulty level was .99.
Figure 5 shows that the misfit items in the area of
the low-difficulty level had improved. The overall
result shows that there was no misfit item that was
typically over or under (infit MSNQ 0.75). We
therefore divided students into three class levels
(i.e., Advanced, Intermediate, and Basic) by
employing a probability model developed using a
naive Bayes classifier.
Figure 5. Result of the placement test ver. 1.0
281
budgetary and manpower savings will thus be

possible.
Regarding the second point, we classified students
into appropriate class levels. The probability
model proposed by our project can be trained and
refined through successive application. With a
more sophisticated probability model, we are
currently working on the development of a naive
Bayes-based computer-adaptive teaching system.
More integrated and sophisticated teaching
support will be offered once the new system
becomes available.
M = 19.04
SD = 6.25
M = 22.41
SD = 5.86
Figure 6. Result of the placement test and the

achievement test
Figure 6 shows the result of 31 shared test items in

the placement test in March, 2013 and the
achievement test in July, 2013. 451 freshmen took
both tests, and we compared the scores of the
shared items. As Figure 6 shows, a tendency
toward improvement of the students vocabulary
and grammar results can be recognized.
In regard to the third point, we managed to

develop an integrated framework for English
language lessons. Using the framework we are
able to identify students English language
abilities and deficits regularly and successively
over time. Data taken from the test are useful not
only for students, but also for tutors and the
university in attempts to improve lessons and the
syllabus. Also, the framework underpins the
minimum standard of students English ability
both in vocabulary and grammar, and can
contribute to the improvement of their learning
outcomes.
4 SUMMARY
As a result of our efforts, we were able to achieve
three project outcomes:
1. Development of an original placement
examination (ver. 1.0)
2. Development of a probability model based on a
naive Bayes classifier
3. Development of an integrated framework for
two-year compulsory English lessons, with
applications to actual lessons
In regard to the first point, it is worth noting we
managed to drastically reduce budgetary and
manpower costs. Beginning in the 2014 academic
year we will conduct a computer-based placement
test using Moodle software, in place of the
previous private, standardized test. Since we have
developed a new Moodle module for a naive
Bayes classifier, scores of any test conducted on
Moodle can be classified automatically. Further
Certain limitations to our efforts did, however,

become clear as we continued our research. The
framework created through the project mainly
focuses on developing and monitoring students
basic English language abilities in areas such as
vocabulary and grammar. Additional efforts are
required to add communicative or practical
support and materials to the framework.
Although limitations were found, we managed to
develop an integrated framework for English
language lessons. The framework endorses the
two-year compulsory English course of study,
from the placement test prior to entrance to the
achievement test in the second year. The system
framework, which enables us to regularly monitor
students English language abilities, provides us
with important data to direct our English
education in an effective manner. With the
globalization of education and the need to address
the decline in students basic English abilities, we
282
are confident the integrated framework we

developed through the project will aid Japanese
universities with similar budgetary, manpower,
and curriculum concerns.
14.
15.
5 REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Japanese Ministry of Education, Culture, Sports,

Science and Technology.: Reformation of Educational
Plan for Development of Global Human Resources
(2013).
http://www.kantei.go.jp/jp/singi/keizaisaisei/skkkaigi/da
i7/siryou07.pdf
Task Force for Developing Global Human Resources.:
Strategy for Developing Global Human Resources
(2012).
http://www.kantei.go.jp/jp/singi/global/suisin/dai2/siryo
u01_1.pdf
Yamamoto, K., Azuma, J., Sumi, S,: Development and
Application of a Blended Learning System for English
Learners. Journal of the University of Marketing and
Distribution Sciences (Humanities, Arts & Sciences) 24
(2), 3337 (2012).
Brown, A. L.: Design Experiments: Theoretical and
Methodological Challenges in Creating Complex
Interventions in Classroom Settings. The Journal of the
Learning Sciences 2 (2), 141178 (1992).
Collins, A.: Toward a Design Science of Education. In:
Scanlon, E., OShea, T. (eds.) New Directions in
Educational Technology, pp. 15-22. Springer, Berlin
(1992).
Bell, P.: On the Theoretical Breadth of Design-based
Research in Education. Educational Psychologist 39 (4),
243253 (2004).
Cobb, P., Confrey, J., diSessa, A., Lehrer, R., Schauble,
L.: Design Experiments in Educational Research.
Educational Researcher 32 (1), 913 (2003).
The Design-Based Research Collective.: Design-based
Research: an Emerging Paradigm for Educational
Inquiry. Educational Researcher 32 (1), 58 (2003).
Cobb, P., et. al.. op. cit. (2003)
Hung, H. T.: Design-based Research: Designing a
Multimedia Environment to Support Language Learning.
Innovations in Education and Teaching International 48
(2), 159169 (2011).
Gravemeijer, K., Cobb, P.: Design Research from the
Learning Design Perspective. In: van den Akker, J.,
Gravemeijer, K., McKenney, S., Nieveen, N. (eds.)
Educational
Design
Research:
The
Design,
Development and Evaluation of Programs, Processes
and Products, pp. 17-51. Routledge, London (2006).
Yutdhana, S.: Design-based Research in CALL. In:
Egbert, J.L., Petrie, G.M. (eds.) CALL Research
Perspectives, pp. 169178. Lawrence Erlbaum
Associates, Mahwah, NJ (2005).
Amiel, T., Reeves, T.C.: Design-based Research and
Educational Technology: Rethinking Technology and
16.
17.
18.
the Research Agenda. Educational Technology &

Society 11 (4), 2940 (2008).
Regional Science and Center for Higher Education.:
(2013) Application of TOEFL Test for Development
of Global Human Resources. 3-013-09.
Sumi, S., Takeuchi, O.: Using an LMS for foreign
teaching/learning: An attempt based on the Cyclic
Model of Learning. The Journal of Information and
System in Education 7, 5966 (2008).
Takeuchi, O.: Mizukara manabu shisei wo
minitukerutameniha. [Autonomy and self-learning:
Necessities and methods] Teaching English Now 8, 25
(2007).
Takeuchi, O.: Gaikokugo shutoku no seihi:
Kyoshitsugai no doryoku ga kagi. [Success or failure in
foreign language learning: Some keys for success]
Mainichi Shinbun (2002).
http://moodle.umds.ac.jp
ACKNOWLEDGMENT
This work was supported by JSPS KAKENHI
Grant Number 25870967.
283
An Active Middleware for Secure Automatic Reconfiguration of Applications for

Android Devices
S. Kami Makki
Department of Computer Science, Lamar University, Texas, USA
kami.makki@lamar.edu
Karim Abdelrazek
Department of Computer Science, The City College of New York, New York, USA
kabdelr00@ccny.cuny.edu
Shui Yu
School of Information Technology, Deakin University, Victoria, Australia
syu@deakin.edu.au
AbstractWith the prevalence of smart phones and
the role they play in the lives of consumers, the demand
for high performing mobile computing is apparent.
Although smartphones today are feature-rich, they are
still resource-scarce; they are limited by their memory,
energy, and processing power. These limitations
constrain the ability of these devices to perform
intensive computational tasks without compromising
the consistency of mobile device performance. As such,
the development of a dynamic and intelligent mobile
middleware solution can ameliorate these constraints
through the utilization of surrogate computing
methodologies. In this paper, we present an intelligent
and active middleware solution for secure automatic
reconfiguration of applications for android devices.
This middleware offers efficiency and enhances the
conservation of resources for these devices.
Index Terms - Application offloading; Context
awareness; Feature models; Mobile middleware;
Reconfiguration; Surrogate computing
I. INTRODUCTION
A smartphone is a mobile device, which offers PClike functionality. A variety of operating systems
exist for smartphones today; examples of this include
Apples iOS, Android, Symbian, and Windows
This material is based upon the work supported by the National Science
Foundation under Grant No. 0851912.
Any opinions, findings, and conclusions or recommendations expressed in this
material are those of the author(s) and do not necessarily reflect the views of the
National Science Foundation.
Mobile. Each of these operating systems allow for

the execution of smartphone applications, which
range from gaming to productivity applications.
These days, smartphones are an integral part of
many peoples lives. As a recent report indicates, the
number of people that have mobile phone is growing
rapidly [1]. Out of the five billion mobile phones
available, approximately one billion smartphones are
in use worldwide. The United States alone has over
91 million smartphone users and it is projected to
grow much more [1].
As users become increasingly dependent on
smartphones for their daily life (e.g. banking, maps,
ticketing, etc.), the need for mobile devices with rich
computing capabilities is apparent. However,
smartphones have several limitations: they have finite
energy resources, contain highly variable network
connections, and are resource-poor compared to
standard desktop computers. With these limitations,
smartphones
generally
cannot
perform
computationally-intensive
tasks
without
compromising mobile performance. Thus, these
constraints lower the quality of interaction between
mobile devices and their users.
Constraints facing mobile devices can be alleviated
in a variety of ways. Manufacturers of mobile devices
produce smartphones with improved phone
specifications, such as faster processors and bigger
batteries. However, generating high-end hardware
for these devices affects manufacturing and device
284
costs while limiting the product in other ways, such
as memory size and weight. While this approach
makes some improvements on mobile device
resources, it still leaves the devices with significant
limitations.
Another
approach
involves
offloading
computational tasks to remote servers in order to
conserve mobile resources. This approach is
generally described as Cyber Foraging, or
application offloading. This approach is a key step
in realizing the pervasive computing vision [21].
Solving the problem of resource constraints on
mobile devices is an important achievement for the
user and for mobile computing. In order to achieve
this, an intelligent and active mobile middleware is
necessary.
The outline of this paper is as follows: In the next
section, we discuss relevant background information
pertaining to surrogate computing, automatic
reconfiguration, remote servers, and mobile
middleware. Section III summarizes and highlights
the key features of our proposed middleware. Section
IV presents the conclusion and future work.
II. RELATED WORK
A. Mobile Applications
A mobile application is a program designed for
mobile devices. There are a plethora of mobile
applications (apps) available today; for instance, the
iPhones App Store is home to over 500,000 mobile
applications [2]. Similarly, Androids Google Play
contains approximately the same amount of mobile
applications [3]. Mobile applications range from
games to navigation apps; each application put
different strains on the resources of a mobile device.
For example, a movie streaming application heavily
consumes networking resources, while a game with
detailed 3D graphics consumes processor resources.
B. Offloading
For many years, researchers have investigated the
issues of conserving mobile resources through the
offloading of computationally intensive tasks.
Rudenko et. al. [4] was among the first to introduce
the idea of remote execution to conserve energy of
mobile devices. The authors in [5, 18, 20, 22]
reintroduced the concept of remote execution in

pervasive environments, coining the term Cyber
Foraging. Cyber Foraging revolves around mobile
devices offloading computational tasks to resourcerich surrogates in order to conserve mobile
resources such as memory, and battery life. This
approach has been investigated [6] and shown to be a
viable solution [6, 7, 8] for minimizing the
consumption of resources of mobile devices. Recent
research revolves around the utilization of cloud
resources for offloading [7, 8, 9, 10].
To show how the surrogate computing can be
utilized easily for computation, we have developed an
application which extends the Google Plays search
functionality. Google Play provides users with access
to hundreds of thousands of applications. However,
it is a challenge for users to find applications that
they may like, since they have been given limited
search options. That is because Google Play only
allows searching for apps by keyword or popular
apps featured, while users may want to search for
apps with different criteria.
Our developed
application gives users the ability to set a variety of
preferences (such as app type, popularity, and cost)
in order to tailor search results to their interests.
Figure 1: User Preference Screen
As shown in Figure 1 each preference is given a

value from 1 to 5, with 5 indicating the highest
priority for that particular preference. Once the user
makes a search query for a particular type of app, this
query is offloaded to a private server. The server
sends the query to Google Play for retrieving the
285
search results. The server calculates a score for each
result based on the user preferences. Then, it
organizes the results so only the apps relevant to the
users interests appear, and finally the server sends
the results to the user. Without using a surrogate,
this process is unacceptably slow, as memory and
processing resources of a mobile device are limited.
Our experiment demonstrates the benefits of
offloading to minimize mobile device resource usage,
highlighting the importance of having automatic
reconfiguration in middleware.
C. Remote Server
In order for tasks or applications to be offloaded to
servers remotely, servers must have the following
features to run effectively:
Communication Manager - A communication
manager is needed to operate data to and
from mobile devices.
Virtual Machine Manager- This section of a
remote server manages the creation,
execution, and storage of the virtual machine
within the server.
Decryption Service A decryption service is
required to decrypt data exchanged between
severs and mobile device.
Individual Server Provides service to Mobile Devices
Communication
Manager
Decryption
Service
Phone OS
Storage
Virtual Machine
Manager
Virtual Machine
iOS
Device
Android
Device
Web OS
Device
Android
Device
Figure 2: Remote Server Design
Figure 2 illustrates the design of such a remote

server [11]. The server houses multiple virtual
machines, with each one executing different
applications for mobile devices. It is important to
note that hosting multiple virtual machines is only
possible if the server is scalable. The operating
system of the virtual machine is determined by the
operating system of the mobile device. As previously
mentioned, the communication manager coordinates

all communication between the mobile device and
remote server; it also accesses the decryption service
to decrypt incoming information, and encrypt
outgoing data.
D. Middleware
Middleware is the management layer that manages
service, monitors access, plans the communication
between local and remote services and efficiently
executes the operations required for switching
services [12]. As such, it is responsible for the
reconfiguration of applications for purposes of
offloading. There are three main types of
reconfigurations [11]:
Static
Reconfiguration
With
static
reconfiguration, applications are always offloaded
whenever they are executed on the mobile device.
This process occurs at the time of installation, and
before execution of the application, allowing the user
to change preferences for static reconfiguration of a
specified application.
Dynamic Reconfiguration Unlike static
reconfiguration, dynamic reconfiguration is a process
of offloading based on the state of the device. This
occurs when there is a noticeable change in
performance. In order to determine when a dynamic
reconfiguration should take place, cost benefit
formulas are used. These formulas take into account
several factors, such as battery level, user
preferences, bandwidth description, and network
strength. The formulas are calculated individually and
compared. If the benefit of offloading is higher than
the cost, the middleware will offload the application
to a remote server.
Contextual Reconfiguration Contextual
reconfiguration shares similarities with Dynamic
configuration, with the exception of where
applications are offloaded. With contextual
reconfiguration, only nearby servers are used to host
applications. The term originates from the use of the
devices context, specifically its location. Utilizing
contextual reconfiguration may reduce the latency of
offloading from the mobile device to the remote
server.
Criteria for Good Middleware
In the context of being able to perform efficient
286
automatic reconfiguration of applications, a good
mobile middleware solution must generally meet one
or more of these five criteria:
i. Complete Service Along with reconfiguration,
middleware must offer other features, such as
fault solution systems and intelligent knowledgebase.
ii. Application
Portability
Middleware
compatibility with applications is essential for
providing efficient automatic reconfiguration.
iii. Offloading Pieces of Application Small
portions of an application or the application itself
may need to be offloaded if they are
computationally intense.
iv. Triggering of Reconfiguration.
v. Contextual Reconfiguration Contextual
reconfiguration allows for data to be sent across
shorter distances, allowing for more efficient
offloading.
E. Offloading
Related work in this area has touched upon
different aspects of mobile middleware development
and mechanisms for offloading.
Xiaohu Gu et. at.s work describes an adaptive
offloading mechanism which relies on a distributed
offloading platform and an offloading inference
engine to compute small portions of an application
remotely [13].
Similarly, Gonzalo Huerta-Canepa et. al., worked
on a version of this offloading technique that
revolved around monitoring application behavior and
offloading based on a mathematical model [15].
Mararasu et al.s work describes a service which
must be created and configured by developers [12].
Byung-Gon Chun et. al., discussed the design and
implementation of the CloneCloud system [7, 8],
which enables mobile applications to seamlessly offload threads from mobile devices onto device clones
operating in a computational cloud. This is done
through the utilization of static analysis and dynamic
profiling to partition applications in a manner which
optimizes application execution. Implementation of
this system was shown to lead to a 20 times
execution speedup for certain applications [8].
III. MIDDLEWARE DESIGN

We base the design for our system on our earlier
design [11]. Our propose system provides an
intelligent mobile middleware solution offering
context-awareness, an intelligent knowledge-base to
handle errors, and automatic reconfiguration of
applications. This middleware strives to give
developers a way to make their applications
reconfigurable for offloading with little application
re-development. This section summarizes how our
active middleware works, as well as, highlights its
key components.
In addition to offering a solution for all the criteria
listed in Section II (sub-section D), our system also
contains the following features:
Transparency - Tasks performed by
middleware should be inconspicuous; the user
should be oblivious to anything but the
content being received on the mobile device.
Modularity - Modularity is necessary in order
to allow for easy modification of the
middleware system.
These features are necessary for a well-developed
middleware
solution,
and
ensure
efficient
performance on mobile devices.
A. Application Compatibility
Applications with our system have three
requirements:
XML Information File - An XML information
file is necessary to tell to our proposed
middleware whether the application is able to
be offloaded and if static reconfiguration is
preferred.
Separation of Interface and Computation
Applications need to have an interface which
interacts with the user and a computation
portion which manipulates data.
Transparent RPC - Transparent Remote
Procedural Calls (RPC) is needed for
transferring data between various portions of
the application. This would utilize
a
serialization format to communicate between
the two portions. When this information is
shared with the application, the application
would translate this data into the data used by
the application.
287
B. Reconfiguration
The proposed middleware supports dynamic,
static, and contextual reconfiguration:
Dynamic Reconfiguration - Our system utilizes
the cost benefit using the formulas shown below:
Cost of offloading = Network Type + Battery Level +
Bandwidth Restrictions + User Preferences+ Current
Network Strength + Application Currently
Reconfigured
(1)
Benefit of offloading = Processor Utilization +
Memory Utilization
(2)
Static Reconfiguration - As mentioned in Section

II, with static reconfiguration the application is
always offloaded. However, this type of
reconfiguration is done only at the users request, or
if a developer places a set of conditional rules for
which static reconfiguration must occur in order for
the application to run successfully on the mobile
device.
Contextual Reconfiguration - With contextual
reconfiguration, applications are only offloaded to
nearby servers. When the user permits contextual
reconfiguration, the procedure that follows is similar
to that of dynamic reconfiguration (e.g., utilizing the
cost-benefit formulas to determine whether
offloading is necessary); however, transferring of the
application would have to occur much faster. The
middleware would seek out available servers nearby,
and then offload the application once one is found.
The process is only triggered when the users move
further away from their current locations.
To demonstrate the performance of our proposed
middleware a tic-tac-toe game application was
developed in Makki et. al., [14]. In this application,
once the middleware has decided that offloading
computation of a program is beneficial to the mobile
device, the middleware searches for a server. After it
locates a remote sever, the middleware pauses the
selected application and offloads it. Meanwhile, the
server creates a virtual machine for the device and
receives the application. After the application is
received, the server executes the application and the
middleware resumes the interface portion of the
application; these two portions then communicate
over transparent RPC.
With this constant
communication, the application has no knowledge

that it has been reconfigured, and continues to run
smoothly.
C. Intelligent Knowledge Base
The middlewares knowledge-base is a fault
repository, a collection of previously encountered
errors along with their solutions. If the middleware
encounters a new problem, it first attempts to use a
relevant solution. If that fails, the middleware will try
to solve the problem through trial and error [11].
When multiple failures are accumulated, they are
recorded into the fault repository, as well as, their
attempted solutions. The knowledge base consists of
three columns: there is one for storage of problem
description (stored as keywords), one for attempted
fixes, and the third is for problem solutions. Figure 3
illustrates how the knowledge base works.
Fault, 1
Fault, n
Solution, 1
Solution, n
Attempts, 1, K
Attempts, n, K
Solution failed. Add fault into knowledge base
Solution, not fixed,

deduce possible
solution
Apply
Solution
Solution
found and
applied.
Fault fixed.
Figure 3. Intelligent Knowledgebase Operation:

K = number of previous faults, n = number of faults
Key Legend
Applies to
Requires Feature
Optional
Implemented
Alternative
Figure 4. Feature Model Key/Legend
D.
Feature Model
The feature model as shown in Figure 4 provides
design information about our proposed system,
which allows us to identify all of the commonalities
that would exist between different versions of the
middleware on different devices [11]. The feature
288
model also allows the middleware to be easily
visualized, as shown in Figures [5-8].
Middleware
Higher
Processing Power
Intelligence
Knowledgebase
Sensor
User
Preferences
Automatic
Reconfiguration
Context
Database
Application
Setup
Figure 5. Architecture Overview

Higher
Processing
Power
Intelligence
Knowledgebase
Error
Knowledgebase
Problems &
Solutions
extremely important in the process of offloading

application. Therefore, a reliable serialization format
is essential for an effective middleware. Sumaray and
Makki, compared the efficiency of data serialization
formatting such as: XML (eXtensible Markup
Language), JSON (JavaScript Object Notation),
ProtoBuf, and Thrift for a mobile platform [16].
They showed that the faster formatting for data
serialization and deserialization is the Binary
formatting (ProtoBuf and Thrift). However, when
taking into consideration other factors such as human
readability and adaptability, JSON proven to be the
best overall data serialization format [16]. Therefore,
we propose to use JSON as the default serialization
format for our proposed middleware. This will
provide the middleware with an efficient means of
data communication, especially in transparent RPC
(discussed earlier). It will also decrease the size of
data being transferred to remote servers, which is
important since smartphones typically have a limited
capability for transferring large amount of data.
Context
Database
Reasoning
Problems
Attempted Fixes &
Solutions
Sensors
Policies
User
Preferences
Figure 6. Intelligence Knowledgebase Feature Model
Our system has two main features: automatic

reconfiguration, and user preferences as shown in
Figure 5. It also has optional features such as
Context Database and Intelligence Knowledgebase as
shown in Figure 5. As shown in Figures [6-8], there
are many alternative sub-features, which together
implement the super-feature, or make up some
portions of it. This highlights the modularity of our
propose system, allowing for the creation of a mobile
middleware suited for several mobile devices.
E. Data Serialization Format
As mentioned before, automatic reconfiguration is
a key component of our intelligent and active mobile
middleware solution. Data communication is
Reasoning
Storage
User
Preferences
Structure
Application
Access
Contextual
Reconfiguration
Robust
API
Figure 7. Context Database Feature Model
F. Fidelity Adaptation
Fidelity Adaption is another approach to
conserving a mobile devices resources. Fidelity
refers to an applications metric of quality [17] that
can be modified to conserve energy resources. For
example, if a user runs a visually intensive application
on his/her mobile device and observes a low frame
rate (because the device cannot handle the intense
computations), then the user can lower the visual
quality of the application so it can run smoothly.
Therefore, the fidelity adaptation approach is ideal
289
as it provides a method of minimizing resource usage
in the absence of surrogate computers. Furthermore,
it can be combined with Cyber Foraging techniques
to optimize mobile device efficiency [18, 19]. For
example, this combination would allow a lower
quality visual application to be offloaded to a
surrogate which leads to a smaller amount of data
being communicated to a remote server. This results
in a lower latency between the server and the mobile
device because smaller data can be parsed quickly.
Middleware
Application
Setup
User
Preferences
Preferences
Storage
User
Interface
Automatic
Reconfiguration
Service
Engine
Static
Reconfiguration
Data
Transfer
Engine
Application
States
State
Collection
Monitoring
Service
Remote Server
Lookup
State
Storage
Policies
If/Else
Engine
necessary to conserve energy of mobile devices.

IV. CONCLUSION
In this paper, we have investigated information
pertaining to middleware technology and surrogate
computing [9]. These technologies improve
conservation of resources of mobile devices and
improve their capability. Therefore, we introduced a
new middleware system, which enables automatic
reconfiguration of applications. It is a complete
mobile middleware system containing key features
such as full application offloading, and an intelligent
knowledgebase
for
fault
tolerance.
What
distinguishes our propose middleware system from
other existing middleware is the utilization of fidelity
adaptation to optimize mobile device resource
conservation. This feature gives mobile devices the
ability to save energy in the absence of surrogate
computers, and also facilitate further conservation of
resources when combined with application
offloading. All these proposed features make our
system a fast, versatile, and developer-friendly
middleware solution.
ACKNOWLEDGMENT
The authors wish to thank Matthew Williamson, a
Lamar University student, who has assisted in
programming tasks of this project.
V. REFERENCES
XML
No
Structure
XML
Encryption
State
Security
Other Data
Encryption
Figure 8. Automatic Reconfiguration and User

Preferences Feature Model
Adding this feature to our system enhances its

capabilities and it can lead to performance gains on
mobile devices. In order to make it work with the
middleware system, the XML information file
described in sub-section A of section III would need
to have additional information specifying whether the
application quality can be adjusted. If it can, then the
middleware will lower its quality and offload it when
[1] Go-Gulf.com, Infographic SMARTPHONE USERS

AROUND THE WORLD STATISTICS AND
FACTS (prepared on 2nd January 2012),
http://www.go-gulf.com/blog/smartphone.
[2] Apple,
Apps
for
iPhone,
http://www.apple.com/iphone/apps-for-iphone/
[3] Google, Android Apps on Google Play,
https://play.google.com/store/apps?hl=en.
[4] Alexey Rudenko and Peter Reiher. Saving Portable
Computer Battery Power through Remote Process
Execution. In Proceedings of Mobile Computing and
Communications Review, Vol.2, No.1, pp.19-26,
1998.
[5] M. Satyanarayanan. Pervasive Computing: Vision and
Challenges, IEEE Personal Communications, Vol.8,
No.4, pp.10-17, 2001.
[6] K. Kumar and Y. H. Lu. Cloud Computing for
Mobile Users: Can Offloading Computation Save
290
Energy. In Proceedings of Computer, Vol.43, No.4,
pp.51-56, April 2010.
[7] C. Byung-Gon and M. Petros. Augmented
Smartphone Applications Through Clone Cloud
Execution. In Proceedings of 12th workshop on Hot
Topics in Operating Systems (HotOS XII), May
2009, Monte Verita, Switzerland.
[8] Byung-Gon Chun, Sunghwan Ihm, Petros Maniatis,
Mayur Naik, and Ashwin Patti. CloneCloud: Elastic
Execution between Mobile Device and Cloud. In
Proceedings of the 6th international conference on
Computer systems (EuroSys '11), pp. 301-314, 2011,
New York, USA, DOI=10.1145/1966445.1966473,
http://doi.acm.org/10.1145/1966445.1966473.
[9] Roelof Kemp, Nicholas Palmer, Thilo Kielmann, and
Henri Bal. The Smartphone and the Cloud: Power to
the User. In Proceedings of the MobiCloud 2010.
[10] Xinwen Zhang, Anugeetha Kunjithapatham, Sangoh
Jeong, and Simon Gibbs. Towards an Elastic
Application Model for Augmenting the Computing
Capabilities of Mobile Devices with Cloud
Computing. In MONET 16(3): pp. 270-284, 2011.
[11] Setfan Ashmore and S. Kami Makki. IMISSAR: An
Intelligent, Mobile Middleware Solution for Secure
Automatic Reconfiguration of Applications, Utilizing
a Feature Model Approach. In Proceedings of the 5th
International Conference on Ubiquitous Information
Management and Communication (ACM SIGKDDSIGAPP ICUIMC 2011), pp. 21-23 February, 2011,
Seoul Korea.
[12] Alin F. Murarasu and Thomas Magendanz. Mobile
Middleware Solution for Automatic Reconfiguration
of Applications. In Proceedings of the 6th International
Conference of Information Technology: New
Generations (ITNG 09), pp.1049-1055, 2009, Las
Vegas USA.
[13] Xiaohui Gu, Alan Messer, Ira Greenberg, Dejan
Milojinic, and Klara Nahrstedt. Adaptive Offloading
for Pervasive Computing. In IEEE Pervasive
Computing Magazine, Vol.3, No.3, pp. 66-73, 2004.
[14] S. Kami Makki, Narasimha B. Srirangam, Venkata
S. Aiswarya, and Shui Yu. Utilizing Intelligent
Middleware for Reconfiguration of Applications on
Android. In Proceedings of International Conference
on Convergence and Hybrid Information Technology
(ICHIT11), pp. 81-89. Springer, Heidelberg,

September 2011, Daejeon, Korea.
[15] Gonzalo Huerta-Cnepa and Dongman Lee. An
Adaptable Application Offloading Scheme based on
Application Behavior. In Proceedings of the workshop
of the 22nd International Conference on Advanced
Information Networking and Applications, pp. 387392, March 2008.
[16] Audie Sumaray and S. Kami Makki. A comparison of
data serialization formats for optimal efficiency on a
mobile platform. In Proceedings of the 6th
International Conference on Ubiquitous Information
Management and Communication (ICUIMC 2012).
ACM, DOI=10.1145/2184751.2184810, .New York,
NY, USA, 2012.
[17] Rajesh Krishna Balan. Powerful Change Part 2:
Reducing the Power Demands of Mobile Devices. In
IEEE Pervasive Computing, Vol. 3, No. 2, pp. 71-73,
IEEE Press, 2004.
[18] Jason Flinn and M. Satyanarayanan. Energy-aware
adaptation for mobile applications. SIGOPS
Operating System Rev. Vol. 34, No. 2, pp. 13-14,
April
2000.
DOI=10.1145/346152.346170
http://doi.acm.org/10.1145/346152.346170.
[19] Eyal DeLara, Dan S. Wallach, and Willy
Zwaenepoel. Puppeteer: Component-based Adaptation
for Mobile Computing. In Proceedings of USITS, pp.
14-25, 2001.
[20] Eduardo Cuervoy, Aruna Balasubramanianz, Dae-ki
Cho, Alec Wolmanx, Stefan Saroiux, Ranveer
Chandrax, and Paramvir Bahlx. MAUI: Making
SmartPhones Last Longer with Code Offload. In
Proceedings of MobiSys, 2010.
[21] M. D. Kristensen. Scavenger: Transparent
Development
of
Efficient
Cyber
Foraging
Applications. In Proceedings of IEEE International
Conference
on
Pervasive
Computing
and
Communications (PerCom 2010), pp.217-226, 2010.
[22] Rajesh Balan, Jason Flinn, M. Satyanarayanan,
Shafeeq Sinnamohideen, and Hen-I Yang. The Case
for Cyber Foraging. In Proceedings of the 10th
workshop on ACM SIGOPS European workshop
(EW 10). New York, NY, USA, pp. 87-92, 2010.
DOI=10.1145/1133373.1133390.
http://doi.acm.org/10.1145/1133373.1133390.
291
Minimization of Mutual Coupling Using Neutralization Line Technique for 2.4 GHz
Wireless Applications
W.N.N.W. Marzudi1, Z.Z. Abidin1, S.Z. Muji1, Ma Yue2 and Raed A. Abd-Alhameed3
1
Research Center of Applied Electromagnetic, Universiti Tun Hussein Onn Malaysia, Johor, Malaysia.
2
National Astronomical Observatories (NAOC), Chinese Academy of Sciences.
3
Mobile and Satellite Communications Research Centre, University of Bradford, United Kingdom
wannoornajwa@gmail.com, zuhairia@uthm.edu.my
ABSTRACT
This paper presented a planar printed multiple-inputmultiple-output (MIMO) antenna with a dimension of
100 x 45 mm2. It composed of two crescent shaped
radiators placed symmetrically with respect to the
ground plane. Neutralization line applied to suppress
mutual coupling. The proposed antenna examined both
theoretically and experimentally, which achieves an
impedance bandwidth of 18.67% (over 2.04-2.46 GHz)
with a reflection coefficient < -10 dB and mutual
coupling minimization of < -20 dB. An evaluation of
MIMO antennas is presented, with analysis of
correlation coefficient, total active reflection
coefficient (TARC), capacity loss and channel
capacity.
These characteristics indicate that the
proposed antenna suitable for some wireless
applications.
KEYWORDS
Multiple-Input-Multiple-Output (MIMO), impedance
bandwidth, mutual coupling, neutralization line
wireless applications.
1 INTRODUCTION
Nowadays and future wireless applications, there
is an insatiability demand for having high data
rate, high link quality, large bandwidth antenna at
both transmitter and receiver. MIMO technologies
have gain much attention recently. The potential
for MIMO antenna systems to improve reliability
and enhance channel capacity in wireless mobile
communications has generated great interest[1]. A
major consideration in MIMO antenna design is to
reduce correlation between the multiple elements,
and
in
particular
the
mutual-coupling
electromagnetic interactions that exist between

multiple elements are significant, because at the
receiver end this effect could largely determine the
performance of the system. Lower mutual
coupling can result in higher antenna efficiencies
and lower correlation coefficients[2]. Significant
research efforts to reduce mutual coupling have
been reported in [3-9]. For example, by
connecting an additional non radiating folded
shorting strip between antenna element and
ground plane [3], port to port isolation lower than
28 dB achieved for lower WLAN band. It is also
interesting to observed that by applying T-shape
slot impedance transformer to both single and dual
band PIFAs, the isolation over 20 dB is obtained
[4]. In [5], the authors proposed the method to
obtained low mutual coupling by cutting two
quarter wavelength slots into the ground plane.
The use of planar soft surfaces proposed in [6] to
reduce mutual coupling. Other methods to reduce
mutual coupling and enhanced isolation of the
MIMO antenna , such as inserting slits on ground
plane [7], I-shaped conductor in modified ground
plane [8] and inserting neutralization line between
antenna element[9] was also promising methods.
In this paper, two-element crescent shaped
MIMO antenna presented for the purpose of
wireless applications. The MIMO antenna consists
of two crescent shaped radiators placed
symmetrically with respect to ground plane with
neutralization line (NL) connected in between of
the two antennas. The total dimensions of this
antenna are 100 x 45 x 1.6 mm3. Both simulated
and measured result of the fabricated prototype
details reported and discussed.
292
2 ANTENNA DESIGN CONCEPTS
(a) Top View
(b) Bottom View
can be observed that the optimal distance of NLd

is at 9.79 mm which gives the lowest mutual
coupling at 2.4 GHz.
Figure 2. Simulated transmission coefficient, S21 of the

various distance location of neutralization line.
Figure 1.Geometry of the proposed antenna (in mm) (a) top

view, and (b) Bottom view.
The proposed antenna geometry is illustrated in

Figure 1. The antenna system comprises with two
crescent shaped radiators that is similar to that in
[10] deployed on an economically FR-4 substrate
with relative permittivity of 4.4 and a thickness of
1.6 mm operating at 2.4 GHz. The radiators are
separated by 0.147 (18.36 mm) for the
minimization of mutual coupling. While, 83 45
mm2 ground plane placed on the other side of the
substrate as shown in Figure 1 (b). The overall
dimensions of the proposed antenna are 100 45
1.6 mm3 which is suitable for wireless
application such as a network card or mobile
device. In addition, in order to improve the mutual
coupling of the proposed antenna, neutralization
line with 0.5 mm width inserted between the
radiators. Theoretically, neutralization lines
transfer some current from the first antenna and
delivers to the second antenna to cancel out the
existing coupling.
To clarify the effectiveness of the
neutralization line (NL) of the proposed antenna,
the parametric study of the location of the
neutralization line, NLd was carried out with the
width of the NL is kept at 0.5 mm. From Fig. 2, it
Figure 3.Comparative plot of s-parameters output for

simulated and measured results for the proposed antenna
without neutralization line.
293
3 SIMULATED AND MEASURED
PERFORMANCE
(over 2.04-2.46 GHz). The bandwidth achieved

fully covered the wireless application such as
network card at 2.4 GHz.
Figure 4.Comparative plot of s-parameters output for

simulated and measured results for the proposed antenna with
neutralization line.
Figure 3 and Figure 4 show the simulated and

measured s-parameters output for the proposed
antenna with and without neutralization line,
respectively. As can been observed, the measured
return loss, |S11| and mutual coupling, |S21| for
both figures (Figure 3 and Figure 4) are
reasonably good agreement with the simulated
results. The resonance frequency is slightly shifted
between the simulated and measured results and
this is probably due to the discrepancy of SMA
connector and fabrication tolerance. It is
apparently seen that by implementing the
neutralization lines, the mutual coupling, |S21| of
the proposed antenna can be improved from -14
dB without neutralization lines to below -21 dB
with neutralization line at 2.4GHz.
To validate the simulated results, the physical
prototypes of the proposed antenna with and
without neutralization line were fabricated and
tested, the top and back view of the proposed
antenna as shown in Figure 5. The S-parameters of
the antenna were measured by Vector Network
Analyser 8722ET (VNA). The measured return
loss |S11| and mutual coupling |S21| are plotted in
Figure 6. As can be seen, when the neutralization
line was inserted, the mutual coupling has been
reduced around 7.14 dB (from -14.63 dB to -21.77
dB) with an impedance bandwidth of 18.67%
(a)
(b)
Figure 5.Practical prototype of the proposed antenna (a) top

view (b) back view.
Figure 6.Measured S-parameters of the proposed antenna

with and without the neutralization line.
294
(a)
The simulated and measured radiation

patterns of the proposed antenna in the x-z plane
(E-plane) and y-z plane (H-plane) with port 1
excited while port 2 terminated with 50 load
plotted in Figure 7 and Figure 8, respectively. The
antenna shows a stable omnidirectional pattern in
the E-plane and H-plane over the operating
frequency of 2.4 GHz.
4 DIVERSITY PERFORMANCES
(b)
Figure 7.Simulated radiation patterns for the proposed
antenna for two planes at 2.4 GHz; (a) X-Z plane.
xxxx (blue) simulated cross-polarization, oooo (red)
simulated co -polarization. (b) Y-Z plane. xxxx (blue)
To evaluate the capabilities of MIMO/diversity

antenna, the envelope correlation coefficient
(ECC) is an important criterion to be presented.
Basically, envelope correlation can be computed
by using S-parameters or radiation pattern of the
antenna. The envelope correlation of the MIMO
antenna system can be expressed by using the
following expression[11]:
simulated co-polarization, oooo (red) simulated crosspolarization Port 1(left) excited and port 2(right)
terminated in 50.
|
e=
11 12
21 22
( 1-| 11 |2 -| 21 |2 )( 1-| 22 |2 -| 12 |2 )
(1)
The simulated envelope correlation coefficient

of the proposed antenna with and without
neutralization showed in Figure 9. An
improvement of the ECC can be seen after the
neutralization line was inserted and it fulfils the
characteristic of diversity pe < 0.5 [7]. Therefore,
the proposed antenna is suitable candidate for
MIMO application.
(a)
Figure 8.Measured radiation (b)

patterns for the proposed
antenna for two planes at 2.4 GHz; (a) X-Z plane. xxxx
(blue) simulated co-polarization, oooo (red) simulated
cross- polarization. (b) Y-Z plane. xxxx (blue) simulated
cross-polarization, oooo (red) simulated co-polarization
Port 1(left) excited and port 2(right) terminated in 50.
Figure 9.Simulated envelope correlation coefficient for the

proposed antenna with and without neutralization line.
295
[
with
(|
]
|
(3)
| )) and
for i , j = 1 or 2.
The simulated capacity losses of the proposed
antenna with and without neutralization line show
in Figure 11. It can be seen that the capacity loss
after inserting neutralization line neither exceeds
0.6 bps / Hz at 2.4 GHz. The comparison between
simulated and measured capacity loss shows in
Figure 12.
(a)
(b)
Figure 10Measured correlation coefficient of the
proposed antenna. (a)without Neutralization line,(b) with
Figure 11.Simulated capacity loss of the proposed antenna

with and without neutralization line.
The measured correlation coefficients of the

proposed antenna with and without neutralization
line are plotted in Figure 10. As can be seen, the
correlation coefficient improved from 0.00042 to
5.162106 after inserting neutralization lines.
The simplified channel capacity loss of a 2 x 2
MIMO system can be evaluated by using the
following equation, given in [9, 12]:
=
Where
matrix:
(2)
is the receiving antenna correlation

Figure 12.Comparison Between simulated and measured
capacity loss of the proposed antenna.
296
Total active reflection coefficient or known as

TARC can be defined as the return loss of the
whole MIMO proposed antenna. Figure 13
showed the simulated total active reflection
coefficient of the proposed antenna with and
without
neutralization
line.
The
better
performance
of
TARC
obtained
with
neutralization line inserted.
Figure 13.Simulated total active reflection coefficient of the

proposed antenna with and without neutralization line.
lower loss of capacity and better performance for

TARC and envelope correlation coefficient
compared to the proposed antenna without
Table 1.Simulated results for correlation coefficient, TARC
and capacity loss at 2.4 GHz
Parameter
Proposed Antenna
without
Neutralization
Line
Proposed Antenna
with
Neutralization
Line
Correlation
Coefficient (dB)
0.091
0.0017
Capacity Loss
(bits/s/Hz)
0.9089
0.6854
TARC (dB)
-12.29
-17.02
In communication system, channel capacity

bounded by the hannons theoretical capacity.
Channel capacity determined as the maximum
amount of information that can be transferred and
received through the channel. Shannon capacity of
the MIMO system can be computed using [13]
2
bits/s/Hz
(4)
Where ;
= ignal to Noise ratio ( NR)
In = Identity matrix of order n
HT = Hermitian transpose of the H matrix
n = no. of antennas
Figure 14. Comparison between simulated and measured

TARC of the proposed antenna.
The channel capacity of the proposed antenna

with and without neutralization line is 5.605
bits/s/Hz and 5.657 bits/s/Hz at 2.4 GHz,
respectively as shown in Figure 15.
The simulated and measured TARC of the

proposed antenna are shown in Figure 14. Small
discrepancies between the simulation and
measurement results can be attributed to
fabrication tolerance and feed cable effects.
The correlation coefficient, TARC and capacity
loss are summarized in Table 1. This shows that
the proposed antenna with neutralization line have
297
[3]
[4]
[5]
Figure 15. Simulated capacity of the proposed antenna with

and without neutralization line.
5 CONCLUSION
A two-element crescent shaped printed MIMO
antenna for covering 2.4 GHz wireless
applications presented. Neutralization line applied
to meet the requirement of MIMO in term of low
mutual coupling parameter. Simulated and
measured results show that the antenna achieves
an impedance bandwidth of 18.67% (over 2.04 2.46 GHz) with a reflection coefficient < -10 dB
and mutual coupling minimization of < -20 dB.
Further, the correlation coefficient, TARC,
capacity and capacity loss have been analyzed for
antenna with and without neutralization line. It has
been shown that the proposed antenna has met the
requirements of MIMO practical antenna.
[6]
[7]
[8]
[9]
[10]
6 ACKNOWLEDGEMENT
The authors of this paper wish to acknowledge the
funding of this project by Universiti Tun Hussein
Onn Malaysia under short term grant Vot 0992.
[11]
[12]
7 REFERENCES
[1]
[2]
G. J. Foschini and M. J. Gans, "On limits of

wireless communications in a fading environment
when using multiple antennas," Wireless personal
communications, vol. 6, pp. 311-335, 1998.
J. Thaysen and K. B. Jakobsen, "Design
considerations for low antenna correlation and
mutual coupling reduction in multi antenna
terminals,"
European
transactions
on
telecommunications, vol. 18, pp. 319-326, 2007.
[13]
H. S. Singh, B. Meruva, G. K. Pandey, P. K. Bharti,

and M. K. Meshram, "Low Mutual Coupling
Between Mimo Antennas By Using Two Folded
Shorting Strips," Progress In Electromagnetics
Research B, vol. 53, 2013.
S. Zhang, B. K. Lau, Y. Tan, Z. Ying, and S. He,
"Mutual coupling reduction of two PIFAs with a Tshape slot impedance transformer for MIMO
mobile terminals," Antennas and Propagation, IEEE
Transactions on, vol. 60, pp. 1521-1531, 2012.
S. Zuo, Y.-Z. Yin, W.-J. Wu, Z.-Y. Zhang, and J.
Ma, "Investigations of reduction of mutual coupling
between two planar monopoles using two /4
slots," Progress In Electromagnetics Research
Letters, vol. 19, pp. 9-18, 2010.
O. Quevedo-Teruel, L. Inclan-Sanchez, and E.
Rajo-Iglesias, "Soft surfaces for reducing mutual
coupling between loaded pifa antennas," Antennas
and Wireless Propagation Letters, IEEE, vol. 9, pp.
91-94, 2010.
J.-F. Li, Q.-X. Chu, and T.-G. Huang, "A compact
wideband MIMO antenna with two novel bent
slits,"
Antennas
and
Propagation,
IEEE
Transactions on, vol. 60, pp. 482-489, 2012.
C. See, R. AbdAlhameed, N. McEwan, S. Jones,
R. Asif, and P. Excell, "Design of a printed
MIMO/diversity monopole antenna for future
generation handheld devices," International Journal
of
RF
and
Microwave
ComputerAided
Engineering, 2013.
C. H. See, R. A. Abd-Alhameed, Z. Z. Abidin, N. J.
McEwan, and P. S. Excell, "Wideband printed
MIMO/diversity
monopole
antenna
for
WiFi/WiMAX applications," Antennas and
Propagation, IEEE Transactions on, vol. 60, pp.
2028-2035, 2012.
C. H. See, R. A. Abd-Alhameed, D. Zhou, T. H.
Lee, and P. S. Excell, "A crescent-shaped
multiband planar monopole antenna for mobile
wireless applications," Antennas and Wireless
Propagation Letters, IEEE, vol. 9, pp. 152-155,
2010.
L. Xiong and P. Gao, "Compact Dual-Band Printed
Diversity
Antenna
for
Wimax/WLAN
Applications," Progress In Electromagnetics
Research C, vol. 32, pp. 151-165, 2012.
S.-W. Su, C.-T. Lee, and F.-S. Chang, "Printed
MIMO-antenna system using neutralization-line
technique for wireless USB-dongle applications,"
Antennas and Propagation, IEEE Transactions on,
vol. 60, pp. 456-463, 2012.
K.J.Babu, G.U Kiran, T.C Kumaraswamy, V.
Harshavardhan and M.V.V Praveen Kumar,
Channel Capacity Calculations of E haped Patch
Antenna MIMO Array, International Journal of
Wireless Communication and Simulation, vol 2, No
1(2010),pp. 105-111.
298
Framework-based data requirements for IT top management
Virpi Hotti
School of Computing
University of Eastern Finland
Finland
virpi.hotti@uef.fi
ABSTRACT
There are several challenges on the adoptions of the
frameworks (e.g., standards). If the main target is semantic interoperability, then data should be taken into
consideration. The main aim of this paper is to identify
some mechanisms that take closer to the semantic interoperability. Therefore, the framework-based lessons
for top management are based on the asset management standard (ISO 55001:2014 where data are one
type of assets) and enterprise architecture metamodel
(TOGAF content metamodel). The results establish
that none of the asset management requirements for top
management mentions data and the attributes of the
data entity do not support the semantics interoperability.
KEYWORDS
data, asset, enterprise architecture, top management
1 INTRODUCTION
Nowadays, there are international standards that
specify requirements for management systems
within the context of the organization. For example, the management standards are implemented to
improve the effectiveness of organizational operations. The management standards provide requirements or guidelines for organizations to develop and systematically manage their policies,
processes and procedures in order to achieve specific objectives [1]. Furthermore, organizations
are adopting Enterprise Architecture (EA) frameworks for improving the interoperability of the
information systems that are used in the production of services. Interoperability is defined to be
ability of the service to interoperate with differ-
ent technical environments, inside and outside of

the organization [2].
Top management shall provide evidence of its
commitment to planning, establishing, implementing, operating, monitoring, reviewing, maintaining, and improving the service management system and the services [3]. However, the IT governance and IT management are unclear [4]. Adaptions of the frameworks (e.g., standards) increase
the understanding of the IT governance or IT
management.
Governance and management, as well as, business
have their own models. However, they have to
take into considerations data which are provided
and consumed by services or are accessed and
updated through services [2]. Therefore, it is important that the top management takes responsibility for the data. In this paper, we present some
mechanisms to make the data transparent and analyzable (Section 2).
We are interested in data and how the frameworks
set requirements for the top management to take
responsibility for data. Because the data are one
type of assets [5], we find out requirements for top
management from the ISO 55001:2014 (asset
management) [6] standard (Section 3). Furthermore, the Annex SL [7] based standard ISO
55001:2014 (asset management) are integrated
with ISO 9001:2008 because over one million
organizations in over 170 countries have implemented ISO 9001:2008 that can be certified by
any organization, regardless of its field of activity
[8].
299
2 ANALYZABLE DATA
Architecture development methods, for example
the TOGAF Architecture Development Method
(ADM), provide tasks and work products (e.g.,
artifacts and deliverables [9]) for creating and
managing architectures within an enterprise or
organization. A content metamodel defines a formal structure in terms to ensure consistency within
the architecture development method (e.g., ADM).
The main terms of the content metamodel are entities and their attributes.
The data entity supports the creation of data
models, which is then extended by the data extension to include the concept of a data component
[2]. Data components are logical and physical
ones that can be governed and deployed into applications and they have own attributes [2].
There are relationships in where the data entity is
a source entity as follows [2]:
Data Entity is processed by Logical
Application Component.
Data Entity is accessed and updated through
Service.
Data Entity decomposes/relates Data Entity.
Data Entity resides within Logical Data
Component.
There are relationships in where the data entity is
a target entity as follows [2]:
Actor supplies/consumes Data Entity.
Logical Application Component operates on
Data Entity.
Service provides/consumes Data Entity.
Logical Data Component encapsulates Data
Entity.
All metamodel entities have the following attributes [2]:

ID. Unique identifier for the architecture
entity.
Name. Brief name of the architecture entity.
Description. Textual description of the
architecture entity.
Category. User-definable categorization
taxonomy for each metamodel entity.
Source. Location from where the information
was collected.
Owner. Owner of the architecture entity.
Data entity has the following own attributes [2]:
Category. The following categories of data
entity apply: Message, Internally Stored
Entity.
Privacy classification. Level of restriction
placed on access to the data.
Retention classification. Level of retention to
be placed on the data.
The attributes of the data entity do not support the
semantics interoperability because they do not
specify the precise meaning or precise semantics
of data elements that are atomic units of data.
Therefore, we illustrate some mechanisms to fulfill the information architecture descriptions.
First, elements of different data are tabulated
(Table 1). The main idea is to find out elements
that are related to several data.
Table 1. Data and data elements
Data
data
ments
Data1
and
ele-
Element1
Elementn
...
Datan
300
Second, the data dictionary based [10] notation is

combined with the data elements as follows (Table
2): mandatory (x), mandatory and the value is one
of several alternative choices ([]) and optional (()).
Table 4. Categorized data elements

Data
data
ments
Data1
Table 2. Mandatory and optional data elements
and
ele-
Element1
and
ele-
Element1
Data1
Elementn
[]
...
Datan
()
()
([])
Finally, we can list data entities using data

elements (Table 3). Furthermore, we will use the
attributes from the TOGAF content metamodel.
Table 3. Data entities

Data
entities
Description
Category
Source
Owner
Element1
...
Elementn
Instead of the spreadsheets, we can use data management tools. However, first we have to understand what we are doing and why? When we explicitly specify our data elements, then we can
specify, for example, the data sources of our performance indicators, as well as, we can make different kind of data analyzes to develop the organizations. For example, the category can be nominal,
ordinal or continuous, if we want to support data
analyzes (Table 4). The categories, nominal (N),
ordinal (O) or continuous (C) can be used with or
instead of mandatory (x), mandatory and the value
is one of several alternative choices ([]) and optional (()). Nominal data (e.g., names) can be
counted, ordinal data (e.g., ratings) can be counted
and ordered, and continuous data (e.g., amounts)
can be counted, ordered and measured.
Datan
Elementn
...
Data
data
ments
(C)
(N)
(O)
Nominal, ordinal and continuous data have different kinds of algorithms [11] or surrogate models
[12] for example to produce summaries or make
predictions.
3 INTEGRATING ASSET MANAGEMENT

SYSTEMS WITH ISO 9001:2008
New management standards and updates are developed based on the generic Annex SL framework [13]. The Annex SL based standards, for
example ISO 55001:2014 (asset management) [7],
contain the following clauses: leadership (5),
planning (6), support (7), operation (8), performance evaluation (9) and improvement (10).
The technical report ISO/IEC 90006:2013 [14]
provides guidelines for the integration of quality
management requirements (ISO 9001:2008) and
service management system requirements
(ISO/IEC 20000-1:2011) [3]. It is figured out
clauses that the IT top management should be taking into consideration of the system management
requirements (Table 5).
301
Table 5. Mapping Annex SL with ISO 9001:2008
Table 6. Mapping Annex SL to ISO 9001:2008
ISO 9001:2008
ISO 9001:2008
4.1 General requirements
5.1 Management commitment
5.4 Planning
5.3 Quality policy
5.2 Customer focus
5.5 Responsibility, authority and communication
7.2 Customer-related processes
5.6 Management review
4.1 General requirements

5.1 Management
commitment
5.2
Customer
focus
5.3 Quality policy
5.4 Planning
5.5 Responsibility, authority and
communication
5.6 Management
review
7.2
Customerrelated processes
ISO/IEC 200001:2011
4.1.1 Management
commitment
4.1.1 Management
commitment
4.1.4 Management
representatives
4.1.2
Service
management policy
4.1.1 Management
commitment
4.1.1 Management
commitment
4.1.2
Service
management policy
4.1.4 Management
representatives
4.5.4.3 Management review
4.1.3 Authority,
responsibility and
communication
4.1.4 Management
representatives
Annex SL
5.1
Leadership
and commitment
5.1
Leadership
and commitment
5.3 Organization
roles, responsibilities and authorities
5.2 Policy
5.1
Leadership
and commitment
5.1
Leadership
and commitment
5.2 Policy
5.3 Organization
9.3. Management
review
5.3 Organization
There are five clauses in ISO/IEC 20000-1:2011

items of which allocates requirements for top
management. The ISO/IEC 20000-1:2011 requirements for top management are mapped to the
Annex SL requirements [15]. When the requirements of Annex SL are mapped to ISO 9001:2008
the technical report ISO/IEC TR 90006:2013 is
used because it has the high level comparison of
ISO 9001:2008 and ISO/IEC 20000-1:2013. Then
we can map Annex SL with ISO 9001:2008
straightforwardly (Table 6).
Annex SL
5.1 Leadership and commitment
5.2 Policy
5.3 Organization roles, responsibilities and authorities
Finally, it is illustrated how the asset management

system requirements (ISO 55001:2014) are integrated with ISO 9001:2008 (Table 7). The requirements are mapped with the mapping Annex
SL with ISO 9001:2008 (Table 5).
None of those 19 requirements for top management mentions data. However, some other of enterprise entities is mentioned (e.g., objective, process and role [2]). We can see importance of data
implicitly, i.e., objectives are tracked against
measures that set performance criteria for services
that consumes/provides data entities [2]. Two
main principles of governance frameworks are
conformance and performance conformance
will be established when the service meets the
requirements which are specified in the contracts;
performance is deemed to be the fulfillment of
the obligations, which are based on contracts or
objectives [4].
302
Table 7. Mapping ISO 55001:2014 to ISO 9001:2008 where

X=asset, M=management, S=system, SDL=shall
demonstrate leadership and commitment with respect to the
XMS, SA=shall assign the responsibility and authority
[6]
ISO 9001:2008
4.1General requirements
5.1Management
commitment
5.4 Planning
5.3 Quality policy
ISO 55001:2014
SDL by ensuring that the
XM policy, the SAMP and
XM
objectives
are
established
and
are
compatible
with
the
organizational
objectives
(5.1)
SDL by ensuring the
integration of the XMS
requirements
into
the
organizations
business
processes (5.1)
resources needed for the
XMS are available (5.1)
SDL by communicating the
importance of effective
XMS and of conforming the
XMS requirements (5.1)
XMS achieves its intended
outcome(s) (5.1)
SDL by directing and
supporting
persons
to
contribute
to
the
effectiveness of XMS
SDL by promoting crossfunctional
collaboration
within the organization (5.1)
SDL
by
promoting
continual improvement (5.1)
SDL by supporting other
relevant management roles
to
demonstrate
their
leadership as it applies to
their areas of responsibility
(5.1)
approach used for managing
risk in asset management is
aligned
with
the
organizations approach for
managing risk (5.1)
shall establish a XM policy
that a) is appropriate to the
purpose
of
the
organization . . . d) includes
a commitment to continual
improvement of the XMS
(5.2)
5.2 Customer focus

5.5 Responsibility, authority and communication
7.2 Customer-related
processes
shall ensure that the

responsibilities
and
authorities for relevant roles
are
assigned
and
communicated with the
organization (5.3)
SA for establishing and
updating
the
SAMP,
including XM objectives
(5.3 a)
SA for ensuring that the
XMS supports delivery of
the SAMP (5.3 b)
SA for ensuring that the
XMS conforms to the
requirements
of
the
International Standard (5.3
c)
SA for ensuring the
suitability, adequacy and
effectiveness of the XMS
(5.3 d)
SA for establishing and
updating the XM plan(s)
(5.3 e)
SA for reporting on the
performance of the XMS to
top management (5.3 f)
shall
review
the
organizations XMS, at
planned intervals, to ensure
its continuing suitability,
adequacy and effectiveness
(9.3)
4 DISCUSSIONS AND CONCLUSION

Nowadays, it is important to be familiar with data
assets. Furthermore, we have to be familiar with
different kind of data ecosystems, and first of all,
what we can do with our data assets. Nominal,
ordinal and continuous data have different kinds
of mechanisms, for example, to produce summaries or make predictions.
There is a jungle of the frameworks for quality
management or improvement, IT governance and
management, project management, as well as,
enterprise management. In this paper, the main
aim was to clarify how the asset management
standard (ISO 55001:2014) and enterprise architecture metamodel (TOGAF content metamodel)
require top management to take data into consid-
303
eration. The results are alarming - none of the asset management requirements for top management
mentions data and the attributes of the data entity
do not support the semantics interoperability.
There is a need for a data-based framework supporting the IT governance, IT management and
enterprise architecture work. The data-based
framework enables dialogue between various
stakeholders. Furthermore, it is needed to ensure
consistency in information architecture supporting
decision-making.
REFERENCES
[1]
ISO/IEC, Consolidated ISO Supplement Procedures

specific to ISO, Directives, Part 1, Fourth edition, 2013,
http://isotc.iso.org/livelink/livelink/fetch/2000/2122/423
0450/4230452/ISO_IEC_Directives%2C_Part_1_and_C
onsolidated_ISO_Supplement__Procedures_specific_to_ISO%2C_4th_edition_2013_
%28PDF_format%29.pdf?nodeid=14883571&vernum=
-2
[2]
The Open Group, TOGAF - Content Metamodel,

http://pubs.opengroup.org/architecture/togaf91doc/arch/chap34.html
[3]
ISO/IEC, Information technology. Service management.

Part 1: Service management system requirements,
International standard ISO/IEC 20000-1:2011
[4]
Lahtela A., Hotti V., Salomaa H.: Service Support in IT

Governance, IT Management and Enterprise
Architecture Context. The Fourth International
Conference on Digital Information Processing and
Communications (ICDIPC2014), SDIWC 2014, 166172.
[5]
ISO/IEC 27013:2012, Information technology -Security techniques -- Guidance on the integrated

implementation of ISO/IEC 27001 and ISO/IEC 200001
[6]
ISO, Asset management Management systems

Requirements, International standard ISO 55001:2014
[7]
The International Register of Certificated Auditors

(IRCA), Annex SL (previously ISO Guide 83),
http://www.irca.org/en-gb/resources/Guidancenotes/Annex-SL-previously-ISO-Guide-83/
[8]
International Organization for Standardization (ISO),

ISO
9000
Quality
management,
http://www.iso.org/iso/iso_9000
[9]
The Open Group, Software & Systems Process

Engineering
Meta-Model
Specification,
http://www.omg.org/spec/SPEM/2.0/
[10]
Structured
analysis
wiki,
Data
Dictionaries,
http://yourdon.com/strucanalysis/wiki/index.php?title=C
hapter_10
[11]
Microsoft, Data Mining Algorithms (Analysis Services Data

Mining),
http://technet.microsoft.com/enus/library/ms175595%28v=sql.105%29.aspx
[12]
Simafore LLC, Pathway,

http://vistasc.simafore.com/vistasc-tree/#
[13]
The International Register of Certificated Auditors

(IRCA), Annex SL (previously ISO Guide 83),
http://www.irca.org/en-gb/resources/Guidancenotes/Annex-SL-previously-ISO-Guide-83/
[14]
ISO/IEC, Information technology Guidelines for the

application of ISO 9001:2008 to IT service management
and its integration with ISO/IEC 20000-1:2011,
Technical report ISO/IEC TR 90006:2013
[15]
Hotti V.: Discipline-free and standard-based

requirements for IT top management. The Fourth
International Conference on Digital Information
Processing and Communications (ICDIPC2014),
SDIWC 2014, 53-59.
304
Literature-based and Architecture-based Modernization

Virpi Hotti and Miia-Maarit Saarelainen
School of Computing
University of Eastern Finland
Finland
virpi.hotti@uef.fi, miia.saarelainen@uef.fi
ABSTRACT
1 INTRODUCTION
Organizational purpose or function changes usually

concern with information systems that have to be
modernized. However, there are difficulties of
making information system modernization decisions
because of suppliers and clients may have different
notions about modernization. In this paper, we
present a review of statements for information
system modernization. The objective of this review
is to describe current reported knowledge in terms
of what kind of modernization is defined and what
the domain of modernization is. We found 42
statements for modernization from 36 papers. These
findings show that the concept of modernization is
unestablished. However, if we want to have similar
notions about information system modernizations
then we have to be transparent where the changes
have effect. The analysis of relationships between
the entities of enterprise architecture is presented to
help in making information system modernization
decisions. The case-based lessons concern the
semantic assets of Finnish social welfare. The
suppliers of three client information systems (CIS)
were evaluated and they answered that 2-58 % of
the semantic assets are unknown. Furthermore,
there are described the main data groups and logical
data stores which are not allowed in the TOGAF
content metamodel.
When something is becoming modern or

something is made modern, it might mean that
something is more suitable for present styles or
needs, or the newer information, methods, or
technology is used [1]. Nowadays, we are
dependent on various information systems. The
information systems can be seen as a collection
of people, processes, data, models, technology
and partly formalized language, which serves
an organizational purpose or function [2]. If the
organizational purpose or function changes,
then we have to make something modern and
quite often we are concerned with information
system modernization.
KEYWORDS
software modernization, system modernization,
architecture-driven modernization, literature review,
enterprise architecture, semantic asset
Many information systems can be called legacy

systems. Legacy systems are typically
described as mission critical systems, systems
that are in an important role in an organization,
and systems which are vital but unmanageable
[3, 4]. Bennett [4] has further described that
legacy systems may contain years of
accumulated experience and knowledge - the
system may be the only place where an
organizations business rules exist. Wallace et
al. [5] cited that system modernization is a
specialized application of system reengineering,
which is the disciplined evolution of a system
from its current state to a new one. Weiderman
et al. [6] present five activities which can be
used with legacy systems: assessment,
maintenance, transformation (can be seen as
reengineering), replacement and combined
strategies. Comella-Dorda et al. [7] divided the
system evolution activities into the following
categories: maintenance, modernization and
305
replacement. Furthermore, they define that

modernization involves more extensive changes
than maintenance and the system must be
replaced when it can no longer be evolved.
In this paper, we are interested in information
system modernization. Our earlier studies have
discussed decision-making in modernizations
[8-10], but during our research we have been
doubtful about the concept of modernization.
Clients and suppliers may have different
notions about modernization [11], which causes
difficulties of making modernization decisions.
Although modernization has been studied from
different perspectives, the concept of
modernization is vague. Our motivation is to
find what modernization is. A definition can be
described as a statement expressing the
essential nature of something [12]. Suonuuti
[13] states: Definitions are statements
describing a concept. In this study, we review
and analyze the statements of information
system modernization. We collect the
statements by using a literature review. In our
review, a statement is something which tells us
what modernization is. Our search returned
altogether 1056 hits and 36 papers included
some kind of a statement or statements of
modernization.
This paper is organized as follows: Section 2
introduces literature-based modernization;
Section 3 presents the architecture-based
modernization efforts in Finnish social welfare.
Section 4 discusses our results.
2 LITERATURE-BASED
MODERNIZATIONS
We are interested in the statements of
information system modernization. Therefore,
we need to find out how the concept of
modernization is defined in scientific papers.
The data collection method in this study is
literature review. Qualitative methods (content
analysis and creating themes) are used for

analyzing the collected material.
Research literature can be reviewed for
different purposes: to provide a theoretical
background for research, to learn the breadth of
the research field or to answer practical
questions by finding out what is said in existing
research literature [14]. In our case, the purpose
for the literature review is to learn what is said
in the existing literature. Our review has been
conducted by adapting two systematic review
guidelines [14, 15]. Okoli and Schabram [14]
describe that a literature review is systematic,
when it 1) follows methodological approach
systematically, 2) explains explicitly how
review has been conducted, 3) includes all
relevant material comprehensively, and 4) can
be
reproduced
by other
researchers.
Kitchenham and Charters [15] define
systematic literature review as a form of
secondary study that uses a well-defined
methodology to identify, analyse and interpret
all available evidence related to a specific
research question in a way that is unbiased and
(to a degree) repeatable. In our review, we
have decided not to assess the quality of the
studies and thus, our review is not systematic in
the means of Kitchenham and Charters [15].
Moreover, in our case, we look for statements
of modernization and we use all the scientific
articles, not only primary studies, from the
computing field which can be retrieved from
the electronic databases we have chosen.
However, we have defined our review process
as follows:
1. Specifying the search terms
2. Selecting the databases
3. Creating the selection criteria
4. Searching the papers
5. Appraising the hits and selecting the
papers
6. Citing the definitions from the papers
Our research strategy is a mix of exploratory
and descriptive strategies. Exploratory strategy
tries to find out what is happening, seeks new
insights and generates new ideas and
306
hypotheses for future research and descriptive

strategy describes the current status of a
phenomenon or a situation [16]. A systematic
literature review aims at identifying, evaluating
and interpreting all available research material
to answer a research question [15]. In our study,
we collected all scientific papers from the field
of computing using the databases we have
access to. We did not use books or grey
material or manual searches. However, we used
all the databases, which were available to us.
After we piloted our search, we decided that the
best way to search for the answer to our
question is to use the search term software
modernization OR system modernization.
We came up this search string, because we
were interested in what is modernized: the
software, the information system, the software
system, the legacy system or the legacy
software. Our search string finds all of these
terms.
We used the following databases to conduct our
search:
ACM Digital Library1
IEEE Xplore2
SpringerLink 3
Elsevier Science Direct4
EBSCOhost Academic Search Elite5
Wiley Online Library6
Emerald7
We did not use manual searches. We included
the papers which are scientific conference or
journal papers, come from the field of
computing, are available in full versions, are
written in English, and include a statement of
(software/system) modernization.
search from other databases. Altogether, we

received 1056 hits from the selected databases.
Finally, after going through all the databases,
we had 36 articles where modernization was
defined (Table 1 ).
Table 1. Searches
Database
ACM Digital Library
IEEE Xplore
SpringerLink
Elsevier Science Direct
Wiley Online Library
Emerald
EBSCOhost
Total
Number
of hits
108
296
228
220
122
18
64
1056
Selected
articles
23
9
2
2
36
With selected articles we started the content

analysis that is the quantification of qualitative
text material [17]. In content analysis the codes
(i.e., sets of words and phrases) are occurrences
of those codes are searched [18]. We found 42
statements that we compared with the
modernization domains (Table 2). In some
papers, there was more than one statement.
In software modernization [19-30] the domain
can be some kind of system and the software
artifacts or assets. In ADM (architecture-driven
modernization) [28, 31-36], the possible
domains are the system, application, software
or software asset. In eight cases modernization
is named divergently: LIS modernization [45],
modernization of enterprise systems [44, 46],
modernizing a legacy system [47, 48], modeldriven modernization [49, 50] and software
system modernization [51].
We excluded papers which are written by the

authors of this paper or are found an earlier
1
http://dl.acm.org/
http://ieeexplore.ieee.org/Xplore/home.jsp
3
http://link.springer.com/
4
http://www.sciencedirect.com/
5
http://www.ebscohost.com/academic/academic-search-elite
6
http://onlinelibrary.wiley.com/
7
http://www.emeraldinsight.com/
2
307
existing software
existing software
artifacts
existing software
assets
existing system
legacy
information
system
legacy software
system modernization
software system modernization
There seem to be several challenges on the

adoptions of the EA frameworks [53, 54]. The
case-based lessons concern the semantic assets
of Finnish social welfare. The data model is an
important part of Finnish social welfares
information architecture and it contains over
200 different social care client records, over
150 core components and over 100
classifications [55]. Actually, the spreadsheetbased core components and social care client
records are nowadays the main semantics assets
of the social welfare in Finland.
1
1
1
1
2
1
1
4
1
1
legacy system
LIS
software system
system
4
total
software modernization
modernization of enterprise
systems
modernizing a legacy system
modernization
model-driven modernization
LIS modernization
Domain
aged legacy
system
aging software
system
enterprise system
existing
applications
existing IT assets
ADM
Defined modernization
Table 2. Comparison of modernization domains and

defined modernizations
1
2
2
4
13
3 ARCHITECTURE-BASED
MODERNIZATION
One-fifth of the public social welfare

organizations in Finland use more than one
client information system (CIS), i.e., they have
different CIS for different services (e.g. child
welfare or disability services) [56]. It is
concluded that the client information systems
do not yet support casework processes, as well
as, the substantial and functional development
of CIS is the most important challenge in the
near future [56]. The casework processes can
be reached by implementing the core
components and client records into different
CIS. Therefore, three CIS, the suppliers of
which are the leaders of the market910 and one
other representative11, were evaluated [57]. The
representatives of the CIS suppliers were
interviewed and their answers were analyzed.
The results are presented numerically, whereas,
the preliminary results
were present
perceptually (Table 3)
Table 3. Analyzed answers
Governments
are
adopting
Enterprise
Architecture (EA) frameworks for improving
the interoperability of the information systems
that are used in the production of services. In
Finland, the EA development framework is
called JHS 1798 the entities of which are
implicitly mapped the entities of the TOGAF
content metamodel [52].
Semantic
assets
Number
Client records
217
Core
components
Core
components in
CIS
148
148
Known
Unsure
Unknown
120-195
(69-93 %)
112-148
(76-100 %)
2-11
(1-5 %)
0-7
(0-5 %)
2-63
(2-29 %)
35-86
(24-58 %)
60-103
(41-70 %)
1-13
(1-9 %)
35-85
(24-58 %)
http://www.cgi.fi/
http://www.tieto.fi/
11
http://www.abilita.fi/index.html
10
http://www.jhs-suositukset.fi/web/guest/jhs/recommendations/179
308
The client records have the name and

description and they are mainly known by the
suppliers. However, the core components are
better
to
known
than
implemented.
Furthermore, all suppliers recognized the same
25 core components, the implementations of
which are not in CIS.
If casework processes are going to be changed,
it is reasonable to be familiar with data entities
such as the core components and client records.
Not only the name and description of the client
record because the client records contain the
specified core components and specific field.
Furthermore, the eWelfare survey [56] did not
take into consideration the challenges of the
core
components
and
client
records
implementations. There are only the results of
the implementation of different classifications
in the client information systems used by public
social service providers [56].
The integrity of the data model is crucial both
in the definition phase and in the
implementation phase.
Adaptations and
explorations of the data model were improved
by the Resource Description Framework (RDF)
for storing and utilizing the Core Component
Technical Specification (CCTS) model [58].
During the years 2013-2015 the data model is
under the inspection, in where the contents of
documents such as the necessity, adequacy,
notation and recurrence of the information are
estimated. However, the RDF-based semantic
assets are not used in the inspection phase
because the versatile know-how for the
interpretation of the RDF-based assets, not to
mention, for the understanding and adapting of
the RDF-based assets is needed [59].
There are other semantic assets of the social
welfare including the information architecture
as follows12:
conceptual model (ksitemalli in
Finnish)
main data groups (ptietoryhmien

kuvaus in Finnish)
dataflow diagram (tietovirtakuvaus in
Finnish)
logical
data
stores
(loogiset
tietovarannot in Finnish)
vocabularies (sanastot ja koodistot in
Finnish)
The conceptual model does not contain any
terms that can be mapped into data entities (i.e.,
social care client records, core components and
classifications). The main data groups and
logical data stores are not related to the data
entities. However, the names of 37 social care
client records are related to processes in the
dataflow diagram. The vocabularies contain the
descriptions of the social care client records13,
core components14 and classifications15.
4 DISCUSSIONS
We noticed that the defined modernization
(system, software, architecture-driven) does not
reveal what the domain of modernization is.
System and software modernization can be
defined similarly as well as software and
architecture-driven modernization. Our findings
reveal the vagueness of the term modernization
[60].
If we want to have similar notions about
information system modernizations, then the
changes have to be transparent. Enterprise
architecture frameworks make the entities and
relationships of the enterprise transparent, and
the information system modernization decisions
can be based on the analysis of relationships
between the entities. We have identified that at
least four entities (i.e., services, data entities,
logical application components, logical
13
12
http://prosessipankki.qpr.com/QPR20122/Portal/QPR.Isapi.dll?QPRPORTAL&*prmav&SES=77T1cGdLYHE
XtUm9u9CqXQ&FMT=p&LAN=fi%u002c1&DTM=&RID=1168941
54286571286
http://www.thl.fi/thl-client/pdfs/a270d99b-58c5-44b0-a3bfcffb01002092
14
http://www.sosiaaliportti.fi/File/3eeffcca-8c86-4a01-a7103963f52eb41d/Tietokomponenttisanasto.pdf
15
http://www.thl.fi/thl-client/pdfs/a270d99b-58c5-44b0-a3bfcffb01002092
309
technology components) should be taken into

account when making the decision. Services
provide governed interfaces to access
organizational functions, as well as, they
provide and consume data entities. Therefore, if
organizational purposes or functions are going
to be changed, it is reasonable to list all data
entities which are provided and consumed by
services to be changed. Furthermore, we have
to be familiar with the logical components of
the information system because the services are
realized through logical application components
and implemented in logical technology
components.
There are several challenges on the adoptions
of the EA frameworks at the government level.
The main target is semantic interoperability.
However, are we closer with semantic
interoperability with numerous architectural
descriptions? The main aim of the paper was
questioning the meaning of several architectural
descriptions. Do they help us to be closer with
semantic interoperability? Furthermore, we
have to questioning the frameworks of the
enterprise architecture. For example, the
attributes of the data entity in the TOGAF
content metamodel do not support the
semantics interoperability because they do not
specify the precise meaning or precise
semantics of data elements that are atomic units
of data.
5 REFERENCES
[1]
Merriam-Webster.com.
,
"modernize",2014,
http://www.merriam-webster.com/dictionary/modernize.
[2] R. Hirschheim, H.K. Klein and K. Lyytinen,
Information systems development and data modeling :
conceptual and philosophical foundations, Cambridge:
Cambridge University Press, 1995, pp. 289.
[3] J. Bisbal, D. Lawless, Bing Wu and J. Grimson,
"Legacy
information
systems:
issues
and
directions," Software, IEEE, vol. 16, pp. 103-111, 1999.
[4] K. Bennett, "Legacy systems: coping
stress," Software, IEEE, vol. 12, pp. 19-23, 1995.
[5] E. Wallace, P.C. Clements and K.C. Wallnau,

"Discovering a system modernization decision
framework: a case study in migrating to distributed
object technology," in Proceedings of the 1996
International Conference on Software Maintenance, pp.
185, 1996.
[6] N. Weiderman, L. Northrop, D. Smith, S. Tilley and
K. Wallnau, "Implications of Distributed Object
Technology for Reengineering," Software Engineering
Institute, Carnegie Mellon University., Tech. Rep.
CMU/SEI-97-TR-005, 1997.
[7] S. Comella-Dorda, K. Wallnau, R. Seacord and J.
Robert, "A Survey of Black-Box Modernization
Approaches for Information Systems," in Proceedings of
the International Conference on Software Maintenance
(ICSM'00), pp. 173, 2000.
[8] M. Saarelainen, J.J. Ahonen, H. Lintinen, J.
Koskinen, I. Kankaanp, H. Sivula, P. Juutilainen and
T. Tilus, "Software modernization and replacement
decision making in industry: a qualitative study," in
Evaluation and Assessment in Software Engineering, pp.
12-21, 2006.
[9] M. Saarelainen, J. Koskinen, J.J. Ahonen, I.
Kankaanp, H. Sivula, H. Lintinen, P. Juutilainen and T.
Tilus, "Group Decision-Making Processes in Industrial
Software Evolution," in International Conference on
Software Engineering Advances, 2007. ICSEA 2007. pp.
78, 2007.
[10] M. Saarelainen, "Why Groups are Used in Software
System Modernization Decisions? Comparing Group
Decision-Making in Private and Public Sector," in 11th
International Conference on Computer and Information
Technology (CIT), 2011 IEEE, pp. 350-355, 2011.
[11] J. Koskinen, H. Lintinen, J.J. Ahonen, T. Tilus and
H. Sivula, "Empirical study of industrial decision making
for software modernizations," in International
Symposium on Empirical Software Engineering, 2005.
pp. 10, 2005.
[12]
Merriam-Webster.com.
,
"definition",2012,
http://www.merriam-webster.com/dictionary/definition.
[13] H. Suonuuti, Guide to Terminology, Helsinki:
Tekniikan sanastokeskus ry, 2001, .
[14] C. Okoli and K. Schabram, "A Guide to Conducting
a Systematic Literature Review of Information Systems
with
310
Research," Sprouts: Working Papers on Information

Systems, vol. 10, 2010.
Proceedings of the 13th International Conference on

Computer Systems and Technologies, pp. 52-57, 2012.
[15] B. Kitchenham and S. Charters, "Guidelines for

performing Systematic Literature Reviews in Software
Engineering," Tech. Rep. EBSE Technical Report EBSE2007-01, 2007.
[26] R. Prez-Castillo, I.G. de Guzmn, M. Piattini and

B. Weber, "Integrating event logs into KDM
repositories," in Proceedings of the 27th Annual ACM
Symposium on Applied Computing, pp. 1095-1102,
2012.
[16] P. Runeson, M. Host, A. Rainer and B. Regnell,

Case Study Research in Software Engineering:
Guidelines and Examples, Hoboken, New Jersey: John
Wiley & Sons, Inc., 2012, .
[17] M.D. Myers, Qualitative Research in Business &
Management, SAGE Publications Ltd, 2009, pp. 296.
[18] J. Eskola and J. Suoranta, Johdatus laadulliseen
tutkimukseen, Tampere: Vastapaino, 2000, pp. 266.
[19] C.T.H. Everaars, F. Arbab and B. Koren, "Using
coordination to restructure sequential source code into a
concurrent program," in IEEE International Conference
on Software Maintenance, 2001. Proceedings. pp. 342351, 2001.
[20] C. Chiang and C. Bayrak, "Legacy Software
Modernization," in IEEE International Conference on
Systems, Man and Cybernetics, 2006. SMC '06. pp.
1304-1309, 2006.
[21] J. Mazn and J. Trujillo, "A model driven
modernization approach for automatically deriving
multidimensional models in data warehouses," in
Conceptual modeling, pp. 56-71, 2007.
[22] Y. Dajsuren, M. Goldstein and D. Moshkovich,
"Modernizing legacy software using a System Grokking
technology," in IEEE International Conference on
Software Maintenance (ICSM), 2010, pp. 1-7, 2010.
[23] R. Prez-Castillo, I.G. de Guzmn and M. Piattini,
"Knowledge Discovery Metamodel-ISO/IEC 19506: A
standard to modernize legacy systems," Computer
Standards & Interfaces, vol. 33, pp. 519-532, 2011.
[24] M. Fernndez-Ropero, R. Prez-Castillo, B. Weber
and M. Piattini, "Empirical assessment of business model
transformations based on model simulation," in
Theory and Practice of Model Transformations, pp. 137151, 2012.
[27] R. Prez-Castillo, J.A. Cruz-Lemus, I.G. de Guzmn

and M. Piattini, "A family of case studies on business
process mining using MARBLE," J.Syst.Software, vol.
85, pp. 1370-1385, 2012.
[28] R. Prez-Castillo, I. Garca-Rodrguez de Guzmn,
I. Caballero and M. Piattini, "Software modernization by
recovering Web services from legacy databases," Journal
of Software: Evolution and Process, vol. 25, pp. 507-533,
2013.
[29] R. Prez-Castillo, I. Garca-Rodrguez de Guzmn,
M. Piattini and S. Places, "A case study on business
process
recovery
using
an
e-government
system," Software: Practice and Experience, vol. 42, pp.
159-189, 2012.
[30] P. Jamshidi, A. Ahmad and C. Pahl, "Cloud
Migration Research: A Systematic Review," IEEE
Transactions on Cloud Computing, vol. 1, pp. 142-157,
2013.
[31] M. Wimmer and G. Kramler, "Bridging
grammarware and modelware," in Proceedings of the
2005 international conference on Satellite Events at the
MoDELS, pp. 159-168, 2006.
[32] R. Prez-Castillo, I.G. de Guzmn, O. Avila-Garca
and M. Piattini, "On the Use of ADM to Contextualize
Data on Legacy Source Code for Software
Modernization," in Proceedings of the 2009 16th
Working Conference on Reverse Engineering, pp. 128132, 2009.
[33] A. Rodrguez, A. Caro and E. Fernndez-Medina,
"Towards framework definition to obtain secure business
process from legacy information systems," in
Proceedings of the first international workshop on Model
driven service engineering and data quality and security,
pp. 17-24, 2009.
[34] J. Canovas and J. Molina, "An Architecture-Driven
Modernization Tool for Calculating Metrics," IEEE
Softw., vol. 27, pp. 37-43, 2010.
[25] K. Normantas, S. Sosunovas and O. Vasilecas, "An

overview of the knowledge discovery meta-model," in
311

"Implementing business process recovery patterns
through QVT transformations," in Proceedings of the
Third international conference on Theory and practice of
model transformations, pp. 168-183, 2010.
"Business
process
archeology
using
MARBLE," Information and Software Technology, vol.
53, pp. 1023-1044, 2011.
[37] J. Koskinen, J.J. Ahonen, H. Sivula, T. Tilus, H.
Lintinen and I. Kankaanp, "Software Modernization
Decision Criteria: An Empirical Study," in Proceedings
of the Ninth European Conference on Software
Maintenance and Reengineering, pp. 324-331, 2005.
[38] E. Putrycz and A.W. Kark, "Recovering business
rules from legacy source code for system modernization,"
in Proceedings of the 2007 international conference on
Advances in rule interchange and applications, pp. 107118, 2007.
[39] S. Chung, J.B.C. An and S. Davalos, "ServiceOriented Software Reengineering: SoSR," in 40th
Annual Hawaii International Conference on System
Sciences, 2007. HICSS 2007. pp. 172c-172c, 2007.
[40] D.J. Meacham, J.B. Michael, M. Shing and J.M.
Voas, "Standards interoperability: Applying software
safety assurance standards to the evolution of legacy
software," in IEEE International Conference on System
of Systems Engineering, 2009. SoSE 2009. pp. 1-8, 2009.
[41] G. Chnard, I. Khriss and A. Salah, "Towards the
Automatic Discovery of Platform Transformation
Templates of Legacy Object-oriented Systems," in
Proceedings of the 6th International Workshop on
Models and Evolution, pp. 51-56, 2012.
[42] E.S. Cho, J.E. Cha and Y.J. Yang, "MARMI-RE: a
method and tools for legacy system modernization," in
Proceedings of the Second international conference on
Software Engineering Research, Management and
Applications, pp. 42-57, 2006.
[43] Z. Li, X. Anming, Z. Naiyue, H. Jianbin and C.
Zhong, "A SOA Modernization Method Based on
Tollgate Model," in Proceedings of the 2009
International Symposium on Information Engineering
and Electronic Commerce, pp. 285-289, 2009.
[44] N. Ilk, J.L. Zhao, P. Goes and P. Hofmann,
"Semantic enrichment process: An approach to software
component
reuse
in
modernizing
enterprise
systems," Inf.Syst.Front., vol. 13, pp. 359-370, jul. 2011.
[45] R. Prez-Castillo, I.G. de Guzmn, M. Piattini, B.
Weber and S. Places, "An empirical comparison of
static and dynamic business process mining," in
Proceedings of the 2011 ACM Symposium on Applied
Computing, pp. 272-279, 2011.
[46] N. Ilk, J.L. Zhao and P. Hofmann, "On Reuse of
Source Code Components in Modernizing Enterprise
Systems," in IEEE Symposium on Advanced
Management of Information for Globalized Enterprises,
2008. AMIGE 2008. pp. 1-5, 2008.
[47] G. Canfora, A.R. Fasolino, G. Frattolillo and P.
Tramontana, "A wrapping approach for migrating legacy
system interactive functionalities to Service Oriented
Architectures," J.Syst.Software, vol. 81, pp. 463-480, 4.
2008.
[48] G.E. Boussaidi, A.B. Belle, S. Vaucher and H. Mili,
"Reconstructing Architectural Views from Legacy
Systems," in 19th Working Conference on Reverse
Engineering (WCRE), 2012, pp. 345-354, 2012.
[49] A. Bergmayr, H. Bruneliere, J.L. Canovas Izquierdo,
J. Gorronogoitia, G. Kousiouris, D. Kyriazis, P. Langer,
A. Menychtas, L. Orue-Echevarria, C. Pezuela and M.
Wimmer, "Migrating Legacy Software to the Cloud with
ARTIST," in 17th European Conference on Software
Maintenance and Reengineering (CSMR), 2013, pp. 465468, 2013.
[50] H. Bagheri and K. Sullivan, "Bottom-up modeldriven development," in 35th International Conference
on Software Engineering (ICSE), 2013, pp. 1221-1224,
2013.
[51] G. Menkhaus and U. Frei, "Legacy system
integration using a grammar-based transformation
system," in 26th International Conference on Information
Technology Interfaces, 2004. pp. 119-124, 2004.
[52] The Open Group. , "TOGAF Version 9.1, an Open
Group Standard".
[53] M. Saarelainen and V. Hotti, Does Enterprise
Architecture Form the Ground for Group Decisions in
eGovernment Programme? Qualitative Study of the
Finnish National Project for IT in Social Services, 15th
IEEE International Enterprise Distributed Object
Computing Conference Workshops (EDOCW), 2011, 1117.
312
[54] D. Makola and V. Hotti, Critical Success Factors

for Adopting Enterprise Architecture Metamodel in the
Health Sector: Literature Review, Journal of Health
Informatics in Africa, 2013.
[55] E. Ailio and J. Krki, Sosiaalihuollon
asiakastietomalli, Finnish Journal of eHealth and
eWelfare, 2013,
http://ojs.tsv.fi/index.php/stty/article/view/8171/6234
[56] P. Hmlinen, J. Reponen, I. Winblad, J. Krki, M.
Laaksonen, H. Hyppnen, and M. Kangas, eHealth and
eWelfare of Finland. Checkpoint 2011, National
Institute for Health and Welfare, Report No.: 5/2013.
[57] H. Lintula, L. Vtinen, T. Ahonen, M. Heiskanen,
V.
Hotti,
Selvitys
muutostarpeista
asiakastietojrjestelmiin
kolmen
tietojrjestelmtoimittajan
nkemyksi
Tikesostavoitetilan
edellyttmist
asiakastietojrjestelmmuutoksista, Sosiaalialan tietoteknilogiahanke, 2011.
[58] K. Hyppnen, M. Alonen, S. Korhonen, and V.
Hotti, XHTML with RDFa as a semantic document
format for CCTS modelled documents and its application
for social services, The Semantic Web: ESWC 2011
Workshops, Lecture Notes in Computer Science, Volume
7117 2012, 229-240, Springer Berlin Heidelberg.
[59] H. Lintula, P. Leinonen, and V. Hotti, Semantic
Documents - Example for CCTS-based and RDFaannotated XHTML documents, The International
Conference on E-Technologies and Business on the Web
(EBW2013), SDIWC 2013, 120-125.
[60] M. Saarelainen and V. Hotti, Different Notions
Lead Difficulties of Making Information System
Modernization Decisions, The Fourth International
Conference on Digital Information Processing and
Communications (ICDIPC2014), SDIWC 2014, 43-52.
313
A Medium Access Protocol for Cognitive Radio Networks

Based on Packet's Collision and Channels' Usage
Dina Tarek Mohamed
d.tarek@fci-cu.edu.eg
Amira M. Kotb
a.kotb@fci-cu.edu.eg
S.H.Ahmed
sana.ola@fci-cu.edu.eg
Information Technology Department. Faculty of Computers and Information. Cairo University

ABSTRACT
In Cognitive Radio Networks (CRNs); the unlicensed
users named Secondary Users (SUs) are allowed to
share the licensed wireless spectrum band with its
licensed users name Primary Users (PUs) but without
degradation in the PU's Quality of Service (QoS). As
all the SUs have the same priority to access the
licensed spectrum band; a spectrum sharing protocol is
needed to fairly divide the available spectrum band
among these SUs. Spectrum sharing protocols in CRNs
are similar to Medium Access Control (MAC)
protocols in regular networks. In this paper; a time
slotted spectrum sharing protocol named Channel
Usage and Collision Based MAC protocol (CUCBMAC) is proposed. It is a modification of a MAC
protocol named Collision-Based MAC protocol (CBMAC), it depends on three parameters for the
allocation of channels to the SUs; 1) counting collisions
number for each SU, 2) predicting the availability
probability for all the available channels and then 3)
excluding some of the available channels. It has been
proved that using the proposed CUCB-MAC protocol
outperforms the original CB-MAC protocol on all the
measured performance metrics.
KEYWORDS
Cognitive Radio Networks - Medium Access Control Spectrum Sharing Time Slotted - Collision - Renewal
Theory
1 INTRODUCTION
In recent years there has been a huge increase
in the access of the limited spectrum for mobile
services, so the fixed spectrum assignment policy
to the unlicensed spectrum began to be inefficient,

in addition to the large portion of the spectrum
assigned to licensed holders or services that
remains unutilized. A new communication
technique named Dynamic Spectrum Access
Network or Cognitive Radio network or Next
Generation (xG) communication network, was
proposed to provide the SUs with the capability to
access the licensed spectrum opportunistically
without any interference with its original users
(Get the Best Available Channel).
CRN is a network that consists of wireless
users, cognitive radio and dynamic spectrum
access capabilities. Based on the intelligence of
cognitive radio, the limited available spectrum and
the unutilized wireless spectrum band, the
cognitive radio network uses these factors to build
a policy to allow its users (SUs) to share and
access licensed wireless spectrum band with its
original users (PUs) without any interference with
these users, and without any degradation in their
quality of service [1,4].
As there exist a large number of SUs and as all
of these SUs have the right to access the available
spectrum band, a spectrum sharing protocol must
exist to coordinate the spectrum access with the
SUs; by providing a scheduling technique to be
able to divide the available spectrum band among
the current SUs. Spectrum sharing is similar to
medium access control (MAC) protocol in the
current networks, but with different challenges for
the cognitive radio networks due to the dynamic
nature of the spectrum and the existence of two
different sets of users; PUs and SUs. Surveys on
MAC techniques for CRNs can be found in [5-10].
314
Even when using the most perfect spectrum
sharing techniques, avoiding collision between
PUs and SUs would be impossible. So; in the
scheduling technique; due to the repeated collision
occurrence, collision statistics (average number of
collisions) must be taken into consideration. Using
this concept a fair scheduling schema was
developed in [11].
A time slotted spectrum sharing protocol is
proposed in this paper, it depends on counting the
number of collisions for each SU, predicting the
ideality of the entire available channels and then
based on the channel usage probability by each
SUs nearby PUs at each time slot; channel
exclusion process is done, before SUs' channel
allocation process is performed. By the exclusion
process; it can be anticipated that collision
occurrence will be decreased when taking into
consideration that any of the available channels
has a high probability to be used by any PU
allocated near by the SU requesting the channel,
this SU won't be given this channel.
The rest of this paper is organized as follows;
various MAC protocols classifications are
explained in section 2. Related work is explained
in section 3. In section 4, the proposed MAC
protocol (CUCB- MAC) will be explained in
details. Evaluation of the proposed protocol will
be given in section 5. Finally summarized
conclusion and future works will be illustrated in
section 6.
2 MEDIUM ACCESS CONTROL
PROTOCOLS CLASSFICATIONS
From various readings it has been found that
there is no single classification for spectrum
sharing techniques in cognitive radio networks. In
fact many classifications exist; some of these
classifications are based on the spectrum access
technique, others are based on architecture, on
spectrum allocation behavior, on scope and on
spectrum access time. These classifications will be
explained in the rest of this section.
MAC protocols in CRNs can be categorized
into three main categories according to the access
method; random access MAC protocols, time
slotted MAC protocols and hybrid MAC protocols

[5,6].
Random access MAC protocols; in these
protocols transmission is done at any time, so
users dont have to be synchronized with the
network. These protocols are mainly based on
Collision Sense Multiple Access with Collision
Avoidance (CSMA/CA).
Time slotted MAC protocols; in these
protocols; time is divided into slots, transmission
must be done in the start of the slot. Users have
to be widely synchronized with the network; this
is done by dividing the time into slots for both
data and control channels.
Hybrid MAC protocols; combine both
previous concepts in many ways. One way is to
be partially time slotted such that the controlling
signals transmission appears at synchronized
time slots, while the data transmission is done at
a random access schema. In another way; data
and control transmissions might be done in a
predefined period of time, using a superframe
which is known to all the users in the network,
during the control and data duration the channel
access might be random.
MAC protocols in CRNs can be also
categorized into two main categories according to
the networks architecture or infrastructure;
centralized MAC protocols and distributed MAC
protocols [1,6,7,8].
Centralized MAC Protocols; in these protocols
there exist a central entity that controls the
spectrum allocation and spectrum access
processes. The sensing itself is distributed
among the SUs in the network, then forwarded
to the central entity that constructs the allocation
map.
Distributed MAC Protocols; in these protocols
the spectrum allocation and spectrum access
processes are done independently by each SU
based on the local or probable global policies.
into two main categories according to spectrum
allocation behavior; cooperative MAC protocols
and non-cooperative MAC protocols [1,6,7,9,10].
315
Cooperative MAC Protocols;
in
these
protocols each nodes interference information is
shared among various nodes. This is usually
done by constructing clusters to locally share the
interference measurements.
Non-Cooperative MAC Protocols; these
protocols are selfish as they don't consider
interference at other nodes; they only consider
the node at hand.
into two main categories according to scope as
intranetwork MAC protocols and internetwork
MAC protocols [7].
Intranetwork MAC Protocols; these protocols

is concerned with spectrum allocation among
SU in a single CRN.
Internetwork MAC Protocols; these protocols
allow various CRNs to perform location and
spectrum overlapping.
into two main categories according to spectrum
access technology; underlay MAC protocols and
overlay MAC protocols [1,6,7,8].
Overlay MAC Protocols; in these protocols
SUs are allowed to access the portion of the
spectrum that is not used by the PUs, so that the
SUs interference with any PU is minimized.
Underlay
MAC Protocols; in these protocols

once the spectrum allocation map are
constructed, the SUs transmit at a certain portion
of the spectrum but with a certain transmission
power that is considered noise to the PUs at that
spectrum portion.
RELATED WORKS
In this section examples for MAC protocols

designed mainly for CRNs will be illustrated.
These protocols will be ordered ascendingly
according to their published date.
A MAC protocol for wireless ad hoc cognitive
radio networks which is based on Time Division
Multiple Access (TDMA) is proposed in [12], it is
named Energy Efficient Multichannel MAC
(ECR-MAC) Protocol. In this protocol it is

assumed that the time domain is partitioned into
beacons with predefined length. Each of these
beacons intervals is split into three windows; Ad
Hoc Traffic Identification Messages (ATIM)
window, sensing window and communication
window. The ATIM window is used for beacons
and to transmit and receive control messages. The
sensing window is used to perform the sensing
process, which is done before the transmission of
data to avoid any collision with any PU. The
communication window is used when a SU
chooses a channel, and it is detected to be idle, this
SU will begin transferring the data in the
communication window. But if the chosen channel
is detected to be busy, this SU will transfer to the
control channel to wait till the next beacon
window.
The protocol proposed in [13] is designed for
synchronous cognitive radio networks; it is a time
slotted protocol that uses hopping sequences to
exchange data. While sending; hopping is done by
the SUs source over the licensed channels based
on the SUs destination pseudo-random hopping
sequence. While receiving, the SUs destination
just switches between channels to get the sent
packets. The same generating algorithm is used by
all the SUs in the network to generate the hopping
sequence. Each SU generates a fixed pseudorandom hopping sequence based on its unique ID
i.e. its MAC address.
Scheduling the spectrum assignment for SUs
in [14] is done based on the knowledge of both the
PUs activity profile and the mobility information
of the SUs. A base station that acts as a central
entity exists in this protocol, where each SU can
directly communicate with it; each of these
communications must be done on a separate
channel at any instant of time to avoid any cochannel interference. Using the activity profile of
the PUs, at each time slot the probability of
existence of each PU is calculated by the base
station. The process of scheduling the spectrum
assignment for SUs is defined formally to be
related to the Maximum Throughput Channel
Scheduling problem (MTCS), in order to obtain
maximum throughput by assigning the appropriate
channel for each SU. The protocol designs a
316
general scheduling framework to solve the MTCS
problem. Maximizing the throughput is expected
to maximize the total Available Transmission
Time (ATT).
A centralized MAC protocol named Fairnessoriented MAC (FMAC) protocol is proposed in

[16]. It deals with the existence of several CRNs.
SUs from the same network perform cooperative
sensing, and then send this information to their
central entity to make the final allocation decision.
Each of these SUs needs to be equipped with two
radios in order to; perform spectrum sensing and
to perform data transmission.
A real-time MAC protocol for cognitive radio

networks is presented in [15], this protocol
provides a deterministic medium access and
opportunistic
spectrum
allocation
for
heterogeneous traffic load. The scheduling in this
MAC protocol is done based on the Earliest
Deadline First (EDF) scheduling technique. The
Earliest Deadline First is a dynamic priority
scheduling technique used when designing a realtime system; it operates by increasing the priority
of the task with shorter deadline, in other words
the task with the earliest deadline is the one that
will be executed first. Also a real-time analysis is
used in this protocol acting as an admission
control tool, so that it only allows traffic that
wont endanger any of the deadlines.
The main idea of the collision based MAC

protocol (CB-MAC) protocol [11] is to make a fair
scheduling among SUs, this is done by not
allowing some SUs to meet more collision than
others. To guarantee such fairness; collision
statistics such as the average number of collisions
is used to prioritized SUs. Incoming SUs packets
are retained in a buffer; so some of them will be
served in the next time slots based on their
priorities.
Secondary Users having packets to send

SU1
All Channels
SU2
SUn
SU3
Central Spectrum
Manager
Channel Status and
PU uses it (if busy)
Nearby PUs for each

SU
Ch2
Number of
Collision for each
SU
Buffer
Ch1
Ch3
Calculate Prob. of
ch. usage for each
PU
Prioritize SU's packets

according to their no.
of collisions
Scheduler
Calculate Estimated
Prob. of ch.
availability
Assign Available Channels to SUs
Chm
At each Slot
Figure 1. Proposed System Architecture (CUCB-MAC)
317
information in its buffer to allocate the
appropriate channel for the SUs packets.
4 CHANNEL USAGE AND COLLISION
BASED MAC (CUCB-MAC) PROTOCOL
The proposed CUCB-MAC protocol is a
centralized time-slotted spectrum sharing protocol
based on counting the number of collisions for
each SU, predicting the ideality of the entire
available channel then excluding some of the
available channels at each time slot.
The process of channel exclusion is performed at
the central spectrum manger. It is done by
performing these steps; first at each sample time
the scheduler has to record status of all the
available channels that are used by any PU, then
the scheduler calculates the channel usage
probability for each channel by each PU. The SU
that is requesting a channel to send a packet must
senses the surrounding area for any existing PUs
and then sends to the scheduler a list contains all
the PUs existing in the surrounding area. So the
SUs -that have a nearby PU with high channel
usage probability for any of the available channels
the scheduler- are prohibited them from using
these channels. Hence the protocol is named;
Channel Usage and Collision Based Medium
Access Control (CUCB-MAC).
Figure 1 shows the architecture of the proposed
CUCB-MAC protocol.
It is obvious from this figure that the proposed
protocol consists of three main parties; SUs,
channels and central spectrum manger.
SUs: represent any of the unlicensed user that
need to send a packet.
Channels: are the divided frequencies the SUs

use to send their packets on.
Central Spectrum Manger: acts as the central

entity that is equipped with both the scheduler
and the buffer. This buffer contains 1)packets
requests sent by the SU, 2)the collected
information from the SUs 3)information about
the channels. The scheduler uses the collected
The CUCB-MAC protocol is divided into two

main phases; one performed as the SUs, and the
second is performed at the central spectrum
manager.
4.1 Phase 1: Performed at Secondary Users
If any SU needs to send a packet: first; it stores
it at its buffer. Second, it senses the surrounding
media to find all the nearby PU. Finally, it sends a
request for the central spectrum manger that
includes; the packet id, list of the nearby PUs and
the number of collisions the packet encounters,
waiting for a channel to be allocated to this packet
by the central spectrum manager at the beginning
of a next slot.
318
Start
no
Has a packet to send

yes
Buffers the packet
New Packet
no
Waiting
Time >
Threshold
yes
Number of Collisions
for this packet =0
yes
no
Drop the
packet
Sense for nearby PUs

Send the request to send this packet to the
central spectrum manager including: * Packet
ID * List of its nearby PUs
* Number of collisions it encounters
no
Sot time
arrives
yes
Central Manager
allocates the SU a
channel
no
each sample time while others are performed at

the beginning of each slot time.
1. It calculates the estimated probability of all
channels availability at each time slot based
on the channel status recorded at each sample.
2. It orders the SUs packets requests based on
each packets encountered number of collision
was sent by the SUs.
3. It calculates the channel usage probability of
all the available channels by each PU based on
the channels statuses recorded at each time
sample, and the list of nearby PUs included in
the SU packet request.
4. It allocates the most appropriate channels for
the SUs with packets requests. This task is
performed with the usage of the output
parameters form the last three tasks.
These tasks will be explained in details in the next
subsections.
yes
The SU uses the allocated channel to
send the waiting packet (after removing it
from the buffer)
Increase no
of Collisions
for this
packet by 1
yes
A PU aims to
use this channel
)
no
The SU finishes
sending this packet
no
)
yes
Stop
Figure 2 Tasks Performed at each SU having a Packet to

Send
Note that if the waiting time of the packet exceeds

a threshold, this packet will be dropped. Figure 2
is a detailed flow chart for the sequence of tasks
that each SU performs whenever it has a packet to
send.
4.2.1 Estimation of Channels Availability

Any algorithm that is used for the prediction of the
channel availability with the usage of the past
channel observations can be used in this step. The
proposed CUCB-MAC protocol uses the renewal
theory process for the prediction of channel
availability [17,18].
Any channel i will be in one of two statuses
either ON or OFF, the ON duration is represented
by a random variable Yi, while the duration of the
OFF period is represented by a random variable
Xi. So the probability distribution function (pdf) of
the ON duration can be
can
, and
be the pdf of the OFF duration. Note that the ON
and OFF durations are supposed to be independent
and identically distribute (iid) random variables.
The arrival of each PU is independent, each of
these transition follow Poisson arrival process.
4.2 Phase 2: Performed At Central Spectrum

Manger
(1)
The central spectrum manger performs the most
dominant tasks that is considered the substantial
tasks in the proposed CUCB-MAC protocol. Some
of these tasks are performed at the beginning of
where > 0 is the parameter of the

distribution
The utilization or the load of channel i is
calculated as follows:
319
(2)
where E[TON] and E[TOFF] are the mean of
the distribution.
E[TON] = 1/Yi and E[TOFF] = 1/Xi
The renewal process is used to calculate i and Xi
as follows:
Assume a vector of ri samples results from the
sensing process of channel i for Tpi duration:
Zi = {Zit1 , Zit2 , Zit2 , ........ Ziti}
(3)
i
where Z tj is the status of the link at time tj
(j = 1, 2, 3, ........ ri), note that status
of a link is either ON (1) or OFF (0)
Four types of transition probability based on ON
(1) or OFF (0); 0 0, 0 1, 1 0, 1 1. With
the usage of the joint probability mass function,
the probability of the four transitions can be
calculated as follows:
(4)
and
The Maximum Likelihood (ML) estimator
can be calculated as follows:
(5)
For each channel the Z value will be changed

based on the channel status recorded during the
sensing process performed by the central spectrum
manger. The ns values will changed based on the
channel status recorded during the sensing process
performed by the central spectrum manger, and
also on the previous channels status.
At each time slot:
The central spectrum manager calculates the
possible probabilities
or
(equation 4) for
(-calculated at
and
each channel) using ,
the time samples using equations 5, 6 and 2
respectively).
4.2.2
Calculation of Probability of Channels

Usage for Primary Users
For each channel the central spectrum manager
has a separate stack with specific predefined
length. This stack is used for reordering the PUs
that previously used the channel for a certain
period of time. In this step the channel status
observations recorded for each channel in the
previous subsection are used. But an extra
checking is done in this process while recording
the channel status, the central spectrum manger
obtain the user ID if the channel is occupied by a
PU. Then this observation is used to update the
stack user of each channel as follows:
(7)
Then; the channel usage probability is calculated

for each PU at each time slot as follows:
(6)
where A = ( i - ( i)2) (ri - 1)
B = - 2A + (ri - 1) - (1 - i) n0 - i n3
C = A - i n0 - (1 - i) n3
n0, n3 denotes the numbers of
occurrences of the 0 0, 1 1
transition respectively
At each time sample:
(8)
4.2.3 Allocation of Channels for Secondary
Users
The central spectrum manger uses the collected
number of collisions encountered from each SU
packet, nearby PU list of each SU, the estimated
320
Start
no
Is there any
waiting requests
yes
no
A new slot
arrives
yes
Drop the packets' request exceeding the
waiting time from the packets' waiting list
Using the number of collisions for each packet:
order the waiting packets requests descendingly.
Using the estimated probability of channels' availability

* Exclude the channels with zero probability.
* Order the channel descendingly based on their availability
yes
Channel list is
empty
no
Pick the packet at the top of the packets' waiting list
Find the SU for this packet then check its nearby PU
Pick the next
channel in the
channel list
Pick the first channel in the channel list

Check the channel usage for each nearby PU
yes
yes
Any of the nearby PU

has usage Proba. for
this channel > Thres.
Any available
channel
no
no
Assign this channel to this packet.
Send the SU a message of that meaning
* Remove the packet from the waiting packets' list
* Remove the channel from the available channel list
Waiting Packets
list is empty
yes
Stop
Using Java a simulation program was built

to develop, evaluate, test and compare the
performance measurements of the proposed
CUCB-MAC protocol with the original CB-MAC
protocol. The built simulation program is divided
into four parts; Packets creation, CB-MAC
protocol, CUCB-MAC protocol and the
calculation of the performance metrics for both
protocols.
To achieve correct results that are comparable
to reality, the packet creation process are done
randomly. The same created packets have to serve
as an input for both protocols to be able to obtain
an accurate comparison between both protocols
performance. Because of that the packet creation
process is a separated function from the two
compared protocols, the output of this creation
process serve as an input for the two MAC
protocols. As for the packet arrival function a
negative exponential distribution [19] is used for
the generation of the packets random arrival time,
associated with random users number.
no
Figure 3. Allocating Available

Channels to Waiting Packets at
the Central Spectrum Manager
probability for each channel availability and the

estimated probability for each channel usage by
each PU; to assign the most appropriate channel
for the waiting SUs packets.
Figure 3 illustrates a flow chart for allocating
available channels to waiting packets done by the
central spectrum manager at each slot.
Table 1. Simulation Parameters
Parameter
Value
PU Data Rate
Number of PUs
SU Data Rate
Number of SU
Packet Length
Number of Channels
Sample Time
Slot Time
Dropping Time
Channel Usage
Threshold for each PU
Simulation Time
20 packets/sec
20
30 packets/sec
20
4500 bytes
4
5 msec
0.6 * Packet Transmission Time
3 * Packet Transmission Time
0.3
10 sec
The used simulation parameters are listed in Table

1 [11,20]. These parameters are changed either up
or down in various scenarios. This is done to test
the proposed protocol upon several constrains and
then prove its efficiency.
5.1 Performance Metrics
5 EVALUATION OF THE PROPOSED

CUCB-MAC PROTOCOL
In this section the performance metrics used to

evaluate and test the proposed CUCB-MAC
protocol are illustrated. Five performance metrics
are used; average throughput, average delay,
321
average number of PUs packets collided,
percentage of channel utilization and finally
percentage of dropped packets.
5.1.1 Average Throughput
Average throughput is defined as the average rate
of packets delivered successfully per one second.
Its preferable to retain it as extreme as possible.
protocol with lower percentage of channel

utilization can send more packets than the other
one. So Its preferable to retain it as minimum as
possible.
5.1.5 Percentage of Dropped Packets
Packets waiting in the SUs buffer for a long time
without being served due to the lack of channels
are useless, so dropped. Its preferable to retain it
as minimum as possible.
(9)
5.1.2 Average Delay
Average Delay is defined as the average time
taken for a packet to be transferred from the
source to the destination. Its preferable to retain it
as minimum as possible.
(10)
5.1.3 Average Number of Primary Users
Packets Collided
Collision occurs when a PU and a SU aim to use
the same channel. Its preferable to retain it as
minimum as possible.
(11)
5.1.4 Percentage of Channel Utilization
Percentage Channel Utilization is defined as the
percentage of how efficient the channels are being
used. Or it can be defined as how much time the
channel is being used from the beginning of the
simulation till its end.
(12)
In reality the aim is to retain the percentage of
channel utilization as extreme as possible, as long
as the packets are transferred safely from the first
attempt. But when using a simulation program;
where the same random packets arrival act as an
input for both the original and the proposed
protocols, the protocol having less percentage of
channel utilization is more efficient. As the
(13)
5.2 Simulation Results
Four scenarios are used to prove the efficiency
of the proposed CUCB-MAC protocol compared
to the original CB-MAC protocol. In each scenario
one simulation parameter takes different values to
be able to see the effect of this change on both
MAC protocols. The first three parameters used in
the first, second and third scenarios are; data rate,
time used to drop a packet and packet length
respectively. The fourth scenario depends on
changing the values of channel usage threshold to
examine its impact on the proposed CUCB-MAC
protocol; of course this parameter wont have any
effect on the original CB-MAC protocol.
5.2.1 First Scenario: Changing Data Rate
In this scenario the effect of changing the SUs
data rate is shown in Figures 4 to 8. The SUs data
rate values vary from 10 packets to 50 packets per
second.
Figure 4 shows that the average throughput of
the proposed CUCB-MAC protocol is higher than
that of the old CB-MAC protocol using various
data rates, the throughput improvement increases
with the increase of the data rate. As increasing
the data rate means increasing number of packets
per second, this will lead to increasing in the
collision probability; so as the need to retransmit
the collided packets, i.e. the time each packet will
take from the source to the destination will
322
increase by the retransmission overhead, so the
number of packets arriving per second
(throughput) will decrease. This cannot be avoided
using CB-MAC protocol, while the proposed
CUCB-MAC protocol prevents some packets with
high collision probability from being transmitted,
so decreases the number of collision and some
packets wont waste time in the retransmission
process; thus retransmission overhead will be
avoided; i.e. the time each packet will take from
the source to the destination will not increase,
leading to an increase in the number of packets
arriving per second (throughput).
Figure 5. Average Delay using Various Data rate
Figure 6 shows that the average number of

collided packets is much lower for the proposed
CUCB-MAC protocol than that of the original
CB-MAC protocol for various data rates. The
difference between the values obtained using the
two protocols highly increase with the increase of
data rate. As an explanation for that, as the data
rate increases, the number of packets per second
increases which may be more than the channels
Figure 4. Average Throughput using Various Data rate
Figure 5 shows that the average delay of the

proposed CUCB-MAC protocol is a bit lower than
that of the original CB-MAC protocol for all used
data rates. Also; it is clear that increasing the data
rate will not increase the difference between the
two averages delay got from applying both
protocols. As increasing the data rate; i.e.
increasing the number of packets per second, will
increase the collision probability, so increase the
packets retransmission process leading to a
retransmission overhead, so increase in the time
taken by a packet to reach its destination (delay).
Using CB-MAC, the number of packets
retransmission is high, so as the delay. While
using CUCB-MAC, some of the retransmission
process obtained using CB-MAC can be avoided,
so lower delay.
Figure 6. Average Number of Packets Collision using

Various Data rate
can bear. Trying to send packets in spite of the

probable happening congestion will lead to
packets' collision. This can sometimes be avoided
using CUCB-MAC, which is not the case using
CB-MAC.
Figure 7 shows that the percentage of channel
utilization of the proposed CUCB-MAC protocol
is lower than that of the original CB-MAC
protocol for various data rates. The difference
between the values of channel utilization resulting
from applying both protocols slightly increases
323
with the increase of data rate. Increasing the data
rate means (increasing number of packets per
second), i.e. increasing in the collision probability;
so the packets' retransmission process will be
performed more frequently. In case of applying
the old CB-MAC protocol the time that each
packet will take from the source to the destination
will increase, so the channels will be used more
often. While using the proposed CUCB-MAC
protocol; the packets with high collision
probability will be prevented from being
transmitted, so it wont use the channels, and they
will remain idle for sometimes.
Figure 8. Percentage of Dropped Packets using Various

Data rate
5.2.2
Figure 7. Percentage of Channel Utilization using Various

Data rate

dropped packets of the proposed CUCB-MAC
protocol is lower than that of the original CBMAC protocol for all used data rates. The
difference between the number of dropped packets
resulting from applying both protocols slightly
increases with the increase of data rate. When the
number of packets per second (data rate)
increases, the probability that packets will collide
increases, so the packets retransmission
probability will increase, a packet may even
needed to be retransmitted more than once, so the
waiting time for these packets will increase. If this
waiting time is too high, the packet will be
dropped, this is the case using CB-MAC. But as
using CUCB-MAC prevents highly probable
packets' collision from happening, this will
decrease the number of packets' retransmission; so
as the packets' waiting time, which will eventually
decrease their probability of being dropped.
Second Scenario: Changing Packet

Dropping Time
Some packets may be worthless to their
destinations, due to the long time they have been
waiting in their SUs buffer; some of these packets
may have collided before even more than once. So
the best decision is to drop these packets instead
of trying to use the spectrum to send a useless
data. In this scenario the effect of changing the
maximum time packets are allowed to wait in their
SUs buffer, on the performance metrics using
both protocols is shown in Figures 9 to 13. The
packet dropping time values varies from 1*packet
transmission time to 5*packet transmission time.
Figure 9 shows that using the proposed
CUCB-MAC protocol leads to higher average
throughput than using the original CB-MAC
protocol for various dropping time. The
improvement of the throughput increases with the
increase of the dropping time. Increasing the
packet dropping time, meaning that the packets
will have longer time to stay in the buffer, so it
will have a larger chance to be served before they
are dropped, this will increase the average
throughput for both protocols. Using CB-MAC;
although most of the packets are served at the end
but they suffer from repeated collision, i.e. longer
waiting time, so take a longer time to successfully
reach their destinations, so the number of packets
delivered per second (throughput) is small. In
contraire; in case of using CUCB-MAC; as the
number of collision is smaller, the waiting time is
324
smaller; so the number of packets delivered per
second (throughput) is higher.
Figure 9. Average Throughput using Various Dropping

Time
Figure 10 shows that for the average delay is

almost the same using both MAC protocols when
the packet dropping time is equal to the packet
transmission time, but as increasing the packet
dropping time the difference between the values
resulting from applying the both protocols differs
by very small values, but still the CUCB-MAC
protocol gives less delay than CB-MAC protocol.
Increasing the packet dropping time, means the
packets can stay in the buffer for much longer
time before it is dropped, so the average delay will
be increased. Although increasing the packet
dropping time will lead to high probability of the
packets being successfully sent; but this may take
a long time. As CB-MAC protocol suffers from
high collision rate than CUCB-MAC protocol; so
higher probability of packets' retransmission; i.e.
longer waiting time, so longer time to successfully
reach the destination (delay).
Figure 10. Average Delay using Various Dropping Time

collided packets of the proposed CUCB-MAC
protocol is lower than that of the original CBMAC protocol while increasing packet dropping
time value. Increasing the packet dropping time,
means that the packet waiting time will increase,
so the packets will be buffered for a longer time,
i.e. the number of the packets staying in the SUs
buffers will increase, so the number of packets
remaining in the networks and needed to be served
increases, and as the spectrum is limited, the
probability that packets collide will increase, some
of these collisions are prevented using CUCBMAC more than those using CB-MAC protocol.

Various Dropping Time
Figure 12 displays the percentage of channel

utilization of both protocols while increasing the
packet dropping time value. It is clear that the
values resulting from applying the proposed
CUCB-MAC protocol is lower than in case of
applying the original CB-MAC protocol. The
packet waiting time will increase when increasing
the packet dropping time, so the packets will stay
for longer time and will have a chance to be sent
and not dropped, this is means that the channels
are being used more and more in other words the
channel utilization increases. As CUCB-MAC
prevents highly probable packets' collision from
happening; in contraire of CB-MAC, so CUCBMAC has lower number of packets'
retransmission; i.e. lower channel utilization.
325
5.2.3
Third Scenario: Changing Packet

Length
In this scenario the effect of changing the SUs
packet length from 1500 to 7500 bytes is shown in
Figures 14 to 18.

Dropping Time

protocol is lower than that of the original CBMAC protocol, while increasing the packet
dropping time value. Increasing the packet
dropping time, increases the packet waiting time,
so the opportunity that the packet will be sent
before it is dropped will increase, i.e. the
percentage of packet dropping will decrease. But
as in CB-MAC, the number of collided packets is
higher than in CUCB-MAC, so packets will stay
longer time in case of using CB-MAC, this time
may reach the maximum allowable waiting time,
so they will be dropped in a higher rate than in
case of using CUCB-MAC.
Noticing from Figure 13 that after a certain
specific of the packet dropping time
approximately 4*packet transmission time, the
percentage of dropped packets is decreasing by a
small value. So the other performance metrics
values either change by a small value or retain a
constant value.
It will be noticed in the next five figures that

for medium packets length (3000, 4500 and 6000
bytes), the CUCB-MAC outperforms the CBMAC. The best performance is always obtained
when the packet length is 4500 bytes, increasing
the packet length above that, will decrease the
performance of both protocols. This because too
long packets will have high probability to collide,
which will lead to bad performance, but still
CUCB-MAC outperforms CB-MAC by a little
value.
Figure 14 shows that using the proposed
CUCB-MAC protocol leads to slightly higher
average throughput than using the original CBMAC protocol for medium packet length.
Increasing the packet length; means increasing the
packet transmission time, which will lead to
increasing the probability of packet being collided
before its transmission. Increasing the number of
collisions will lead to more retransmission so
more retransmissions' overhead, i.e. more time to
transmit a packet successfully, leading to low
throughput. So it is more reasonable to use the
proposed CUCB-MAC protocol that prevents
highly probable packets' collision from happening,
which will lead to higher throughput
Figure 14. Average Throughput using Various Packet

Length
Dropping Time
326
Figure 15 shows that the average delay of
the proposed CUCB-MAC protocol is a bit lower
than that of the original CB-MAC protocol for all
used packet lengths. Also; it is clear that
increasing the packet length will not increase the
difference between the two averages delay got
from applying both protocols. As increasing the
packet length; will increase the collision
probability, so increase the packets retransmission
process which leads to a retransmission overhead,
so increase in the time taken by a packet to reach
its destination. Using CUCB-MAC, some of the
retransmission process obtained using CB-MAC
will be avoided, so this is will lead to lower delay.

Various Packet Length

is a bit lower than that of the original CB-MAC
protocol while varying the packet length.
Increasing the packet length will lead to increasing
transmission time, i.e. increasing in the collision
probability; so the packets' retransmission process
will be needed more frequently. Using CUCBMAC prevents some possible collision from
happening; this means preventing more packets'
retransmission and decreasing the overhead
resulted from this process, which leads to lower
channel utilization than CB-MAC protocol.
Figure 15. Average Delay using Various Packet Length

collided packets is much lower using the proposed
CUCB-MAC protocol than when using the
original CB-MAC protocol for various packet
length. The difference between the values
obtained when using both protocols highly
increase with the increase of packet length, till a
medium value of 4500 bytes then the difference
starts to decrease. This is because increasing the
packet length, will lead to increasing the packet
transmission time, which will increase the
probability of a packet to be collided. So it is
better to use the proposed CUCB-MAC protocol
that avoids some collisions.

Packet Length
Figure 18 shows that the percentage of

dropped packets using the proposed CUCB-MAC
protocol is lower than when using of the original
CB-MAC protocol for medium packet lengths. Of
course increasing the packet length will increase
the probability of collision, i.e. increase the needs
for packet retransmission (which may be done
even more than once), so the waiting time for
327
these packets will increase. But as using CUCBMAC protocol prevents highly probable packets'
collision from happening, this will decrease its
number of packets' retransmission, and the
retransmission process overhead, which will
eventually decrease their probability of being
dropped.
number of packets arriving

(throughput) will decrease.
per
second
Figure 19. Average Throughput using Various Channel

Usage Threshold

Packet Length
5.2.4
Fourth Scenario: Changing Channel

Usage Threshold
One of the parameters used in the channel
allocation process protocol for the proposed
CUCB-MAC is the channel usage probability.
This parameter represents the probability that a
PU is expected to use a specific channel, it acts as
a threshold in the proposed protocol; if the
probability of channel usage by a PU exceeds this
specific threshold, it means that this channel is
expected to be used by a specific PU within the
SUs range waiting for a free channel, so this
channel is not allocated to this SU.
It is clear from Figure 19 that the average
throughput of the proposed CUCB-MAC protocol
is higher than that of the original CB-MAC
protocol while increasing the channel usage
probability threshold. Increasing the channel usage
probability threshold used in CUCB-MAC, will
allow packets with high probable collision to be
sent, leading to collision occurrence so packet
retransmission occurs, i.e. the time each packet
will take from the source to the destination will
increase by the retransmission overhead, so the
Figure 20 shows that the average delay of the

proposed CUCB-MAC protocol is lower than that
of the original CB-MAC protocol while increasing
the channel usage probability threshold. It is clear
that increasing the channel usage probability
threshold used in CUCB-MAC; will allow more
and more packets with high probable collision to
be sent; so increase the collision probability, i.e.
increase the packets' retransmission process
leading to a retransmission overhead, so increase
Figure 20. Average Delay using Various Channel Usage

Threshold
in the time taken by a packet to reach its

destination (delay).
collided packets of the proposed CUCB-MAC
protocol is a lower than that of the original CBMAC protocol for all values of channel usage
328
probability threshold. Increasing the channel
usage probability threshold used in CUCB-MAC,
will allow packets with high probable collision to
be sent; so collision occurs, i.e. more packets to be
resent. The leads to the existence of larger number
of packets in the network, and all of them need to
use the spectrum, i.e. crowded network, so more
and more collision.
Figure 23 shows that the percentage of

protocol is lower than that of the original CBMAC protocol, for all the values used for the
channel usage probability threshold. Increasing
the threshold channel usage probability used in
CUCB-MAC; will allow more packets to be sent
(those that needed to be prevented), so the number
of packets that wants to use the spectrum is
increases and the network is more crowded, so
more packets will be waiting in the buffer, and
more packets will be dropped for exceeding their
waiting time. In other word the percentage of
dropped packets will increase.

Various Channel Usage Threshold

is lower than that of the original CB-MAC
protocol for various channel usage probability

Channel Usage Threshold
5.3 Summarized Results
threshold. The increase of the channel usage

probability threshold will increase the number of
packets opportunity to be sent; rather than to be
dropped; which will lead to increasing the number
of packets using the channels, in other words the
percentage of channel utilization will increase.
Table 2: Percentage of Improvement using the Proposed

CUCB-MAC Protocol Compared to the Original CB-MAC
Changed Parameter

Channel Usage Threshold
The results of applying both the old and the

proposed MAC protocols on various performance
metrics using the three scenarios explained in
section 5.2 are summarized in table 2. The average
values for all performance metrics resulted from
applying both protocols, together with the
percentage of improvement obtained using
CUCB-MAC are put in the table.
Performance Metric
Average
Average
Average
Percentage
value
value
of
Obtained
Obtained
improveme
using
using CBnt Using
CUCBMAC
CUCBMAC
MAC
329
Data Rate
Average Throughput (pkts/sec)
17.31
19.32
11.6%
Average Delay (msec)
124.45
117.14
5.9%
Average No. of Packets

collision
0.52
0.16
69.2%
Perc. of Channel Utilization
79.19
75.94
4.1%
Perc. of Dropped packets
21.21
17.64
16.8%
18.11
20.88
15.3%
140.76
132.50
5.9%

collision
0.61
0.19
68.7%
85.34
81.85
4.1%
22.53
17.13
24.0%
18.57
19.98
7.6%
151.51
147.87
2.4%

collision
0.38
0.13
66.8%
72.77
70.45
3.2%
21.61
18.87
12.7%
20.33
21.62
6.3%
125.34
119.91
4.3%

collision
0.63
0.36
42.4%
86.03
83.88
2.5%
17.22
14.58
15.3%
Dropping Time
Channel Usage Thres.
Packet Length
From table 5.2 it can be noticed that the

average values of the performance metrics when
applying the proposed CUCB-MAC protocol
defeat those when applying the original CB-MAC
protocol in all cases. Meaning that using CUCBMAC improves all the performance metrics'
values. The improvements are medium on the
average throughput, the average delay and the
percentage of channel utilization. While high
improvements are obtained for the percentage of
dropped packets and the average number of
collided packets. As an important note we can
notice the high improvement on the average
number of collided PUs' packets. This is the most
important metric as it affects the PU directly,
which is a main aim for any cognitive radio
network not to degrade the PUs quality of service.
Compared to the CB-MAC protocol, the

proposed CUCB-MAC has some advantages in
addition to those mentioned above: in the CBMAC protocol, the packets are stored in the
central control module buffer. If the buffer is
overloaded, the packets are dropped. This is not
the case in the proposed CUCB-MAC, the packets
are stored each SUs buffer.
Also compared to the CB-MAC protocol, the
proposed CUCB-MAC has some drawbacks: As
any SU with a packet to send (in the proposed
CUCB-MAC), needs to sense its surrounding area
for the existing PUs in its range, then send a list of
these PUs to the central control module together
with the sent request. The central control module
uses these lists to calculate each channel usage by
each SUs nearby PUs. This process leads to the
following:
Consuming some energy of the SUs' battery.
Extra buffering needed to store the list of PUs.
Extra processing time needed by the central
control module to calculate the channel usage,
although it is small.
6 CONCLUSION AND FUTURE WORK
In this paper; a spectrum sharing technique for
cognitive radio networks is proposed. This
protocol depends on three parameters for the
allocation of the channels to various SUs. Those
parameters are: the number of collisions each
packet encounters, the estimated probability for
channel availability and the estimated probability
of channel usage by each PU. So the proposed
MAC protocol is named CUCB-MAC. A
simulation program was built for comparing the
proposed protocol to an old one named CB-MAC
that uses only the first two parameters for the
allocation process. The evaluation of the proposed
CUCB-MAC protocol is done with the usage of
three different scenarios. For each scenario a
simulation parameter is changed either up or down
to examine the impact of this change on both the
proposed CUCB-MAC protocol and the original
CB-MAC protocol.
330
From the results, it is concluded that the
proposed CUCB-MAC protocol overcome the
original CB-MAC protocol on all the measured
performance metrics; average throughput, average
delay, average number of PUs packets collided,
percentage of channel utilization and finally
percentage of dropped packets.
As future work, 1) Formulate a mathematical
model to prove the values used as slot period,
channel usage threshold and packet dropping time.
2)
Propose a decentralized MAC protocol that uses
collision statistics. 3)Use another collision statistics
measure. 4) Use a soft computing paradigm based
on fitness optimization function to reach optimal
best channel allocation process.
7 REFERENCES
a Survey, Computer Networks Journal, vol. 50, no. 13,

pp. 2127-2159, September 2006.
[8] T. Vamsi Krishna and Amitabha Das, "A Survey on
MAC Protocols in OSA Networks", Computer
Networks, vol. 53, no.9 , pp. 1377-1394, June 2009.
[9] Varaka Uday Kanth, Kolli Ravi Chandra and Rayala
Ravi Kumar, "Spectrum Sharing In Cognitive Radio
Networks", International Journal of Engineering Trends
and Technology (IJETT), vol.4, no.4, pp. 1172- 1175,
April 2013.
[10] Badr Benmammar, Asma Amraoui and Francine Krief,
"A Survey on Dynamic Spectrum Access Techniques in
Cognitive Radio Networks", International Journal of
Communication Networks and Information Security
(IJCNIS), vol.5, no. 2, pp. 68-79 , August 2013.
[11] Wanbin Tang, Jing Zhou, Huogen Yu & Shaoqian Li,
"A fair scheduling scheme based on collision statistics
for cognitive radio networks", Concurrency and
Computation: Practice and Experience, vol. 25, no.9,
pp. 1091-1100, June 2012.
[1] Ian Fuat Akyildiz, Won-Yoel Lee, Mehmet Can Vuran

& Shantidev Mohanty, Next Generation/ Dynamic
Spectrum Access/ Cognitive Radio Wireless Networks:
a Survey, Computer Networks Journal, vol. 50, no. 13,
pp. 2127-2159, September 2006.
[12] S. M. Kamruzzaman, "An Energy Efficient

Multichannel MAC Protocol for Cognitive Radio Ad
Hoc
Networks",
International
Journal
of
Communication Networks and Information Security
(IJCNIS), vol. 2, no.2, pp. 112-119, August 2010.
[2] Bin Le, "Building a Cognitive Radio: From Architecture

Definition to Prototype Implementation", Doctoral
Thesis, Faculty of the Virginia Polytechnic Institute and
State University, Virginia, USA, June 2007.
[13] Hang Su, "Design And Analysis of Opportunistic MAC

Protocols For Cognitive Radio Wireless Networks",
PhD Thesis, Texas A&M University, USA, December
2010.
[3] Kwang-Cheng Chen, You-Jyun Peng, Neeli Rashmi

Prasad, Ying-Chang Liang & Sumei Sun, "Cognitive
Radio Network Architecture: Part I - General Structure",
ACM 2nd International Conference on Ubiquitous
Information
Management
and
Communication
(ICUIMC), Suwon, Korea, January - February 2008.
[14] Li Zhang, Kai Zeng & Prasant Mohapatra,

"Opportunistic Spectrum Scheduling for Mobile
Cognitive Radio Networks in White Space", IEEE
Wireless Communications and Networking Conference
(WCNC), Cancun, Mexico, pp. 844-849, March 2011.
[4] Ahmad Mohi-Eldeen Rateb, "Introduction to Cognitive

Radio Systems", Technical report, University of
Technology,
Malaysia,
2008.
http://trg.fke.utm.my/members/ahmad/ introtoCR.pdf,
Last accessed February 2014.
[5] Claudia Cormio and Kaushik R. Chowdhury, "A Survey
on MAC Protocols for Cognitive Radio Networks", Ad
Hoc Networks, no. 7, vol.7, pp.13151329, September
2009.
[6] Nhan Nguyen-Thanh, Anh T. Pham and Van-Tam
Nguyen, "Medium Access Control Design for Cognitive
Radio Networks: A Survey", IEICE Transactions on
Communications, vol.E97-B, no.2, pp. 359-374,
February 2014.
[7] Ian F. Akyildiz, Won-Yoel Lee, Mehmet C. Vuran and
Shantidev. Mohanty, Next Generation / Dynamic
Spectrum Access/ Cognitive Radio Wireless Networks:
[15] Kristina Kunert, Magnus Jonsson and Urban Bilstrup,

"Deterministic real-time medium access for cognitive
industrial radio networks", 9th International Workshop
on Factory Communication Systems (WFCS 2012),
Lemgo/Detmold, Germany, pp. 91-94, May 2012.
[16] Yanxiao Zhao, Min Song and ChunSheng Xin, "
FMAC: A Fair MAC Protocol for Coexisting Cognitive
Radio Networks ", IEEE International Conference on
Computer Communications (IEEE INFOCOM), Turin,
Italy, pp.1474-1482, April 2013.
[17] Hyoil Kim & Kang Geun Shin, "Adaptive MAC-layer
Sensing of Spectrum Availability in Cognitive Radio
Networks", Technical Report, University of Michigan,
2006.
[18] Hyoil Kim & Kang Geun Shin, "Efficient Discovery of
Spectrum Opportunities with MAC Layer Sensing in
Cognitive Radio Networks", IEEE Transaction on
Mobile Computing, vol. 7, no. 5, pp. 533-545, May
2008.
331
[19] Michael Todinov, "Reliability and Risk Models", John
Wiley & Sons, 2005
[20] Xiaoke Jiang & Jun Bi, "What Would Be the Optimal
Chunk Size for Information", Technical Report,
Institute for Network Sciences and Cyberspace,
Tsinghua University. October 2013. http://netarchlab.
tsinghua.edu.cn/~shock/THU-NetArchLab-ICN-TRCHUNK-SIZE-20131101.pdf. Last Accessed February
2014
332
ICT governance: Literature review and Survey in Northern Savonia

Virpi Hottia, Heikki Merilinena, Harri Salomaaa, Antti Lahtelab
a
School of Computing, University of Eastern Finland, Finland
b
Regional State Administrative Agency for Eastern Finland, Finland
virpi.hotti@uef.fi, hmerilai@kapsi.fi, harris@student.uef.fi, antti.lahtela@avi.fi
ABSTRACT
1 INTRODUCTION
IT decision making is supported by frameworks of

the different kind for IT governance, IT management, and enterprise architecture. Organizations
are adopting Enterprise Architecture (EA) frameworks for improving the interoperability of the
information systems that are used in the production of services. Therefore, the main aim of our
literature review (212 hits, 20 appraised hits) was
to find out the latest scientific researches around
IT governance, IT management and IT service
management within enterprise architecture. It is
possible to archive the commitment to things of
the different kind with assessments. Therefore, the
commitment to the enterprise architecture work
and ICT governance is established in spring of
2014 in the Northern Savonia with the online survey. Our survey of ICT governance has 25 questions. It had 331 potential respondents and we got
136 answers. In this paper, we reported the answers of five questions that position the ICT governance in the Northern Savonia.
Nowadays, there are international standards

that specify requirements for management
systems within the context of the organization. For example, the management standards
are implemented to improve the effectiveness
of organizational operations. The management standards provide requirements or
guidelines for organizations to develop and
systematically manage their policies, processes and procedures in order to achieve specific
objectives [1]. Furthermore, organizations
are adopting Enterprise Architecture (EA)
frameworks for improving the interoperability
of the information systems that are used in the
production of services. Interoperability is defined to be ability of the service to interoperate with different technical environments,
inside and outside of the organization [2].
KEYWORDS
ICT governance, enterprise architecture, IT management, IT service management
Governance and management, as well as,

business have their own models. It is important to show where IT should be used and
what it should produce (evaluate), to whose
responsibility the carrying is given (direct)
and to follow that the intended advantages
will be reached (monitor) [3]. It is talk about
governance function, the tasks of which are
based on the EDM model (E=Evaluate,
D=Direct, M=Monitor). For example, the
Corporate governance of information technology (ISO/IEC 38500) [4] models the corporate governance of IT and it divides into
governance function and management function, the tasks of which are based on the
PDCA model (P=Plan, D=Do, C=Check,
A=Act).
333
The main aim of frameworks of the different

kind for IT governance, IT management and
enterprise architecture is to support IT decision making. Our literature review (Section 2)
tries to find out the latest scientific researches
around IT governance, IT management and IT
service management within enterprise architecture.
In Finland, Act on Information Management
Governance in Public Administration [5]
came into force in 20111, and it imposes obligations on how ICT governance must be designed and described in the organization of
public administration. The law forces the public organizations to make enterprise architecture descriptions and manage them. Therefore, the public sector has desired to follow
the instructions of the ICT Standard for Management [6, 7]. However, there seems to be
difficulties to utilize both enterprise architecture and ICT governance that is based on the
ICT Standard for Management2. Our ICT
governance survey (Section 3) is conducted in
the Northern Savonia in spring of 2014. The
survey supports the commitment to the enterprise architecture work and ICT governance.
2 LITERATURE REVIEW
The data collection method in this study is
litera-ture review. Instead of ICT we used IT
because there are more hits using IT. We reviewed IT governance together with IT management, service management and enterprise
architecture. We tried to find frame-works for
clarifying what the service support should be
considered and why. Furthermore, we collected statements for IT governance if it was defined in the selected papers - definitions are
statements describing a concept [8].
Research literature can be reviewed for different purposes: to provide a theoretical background for research, to learn the breadth of
the research field or to answer practical ques-
tions by finding out what is said in existing

research literature [9]. Our main aim is to find
out the latest scientific researches around IT
governance and IT manage-ment, first of all,
to learn the breadth of research on a topic of
interest [9].
Our review has been conducted by adapting
two systematic review guidelines [9, 10], and
our review process is as follows:
1. Specifying the search terms.
2. Selecting the databases from the field of
computing.
3. Searching the papers.
4. Creating the inclusion criteria.
5. Appraising the hits and selecting the
papers.
6. Citing the statements from the papers.
We used IEEE Xplore as a pilot database and
we tried different search terms, for example,
(Title:IT governance AND Abstract:IT management) or (Title:IT governance AND Abstract:service man-agement). Finally, we
chose the databases, the advanced search of
which allow searching from titles and abstracts. We selected the following databases
from the fields of computing and information
technology:
ACM Digital Library3
IEEE Xplore4
Elsevier Science Direct 5
Wiley Online Library6
EBSCOhost Academic Search Elite7
First, we listed all hits (Table 1). In the result
ta-bles we use the following abbreviations:
ITG=IT governance, ITM=IT management,
SM=service management, and EA=enterprise
architecture. Secondly, we included (Table 2)
the papers which are available in full versions, written in English, and published in
scientific conference or journal papers.
http://dl.acm.org/
http://ieeexplore.ieee.org/Xplore/home.jsp
5
http://www.sciencedirect.com/
6
http://onlinelibrary.wiley.com/
7
http://www.ebscohost.com/academic/academic-search-elite
4
http://www.vm.fi/vm/en/04_publications_and_documents/03_docu
ments/20110902ActonI/Tietohallintolaki_englanniksi.pdf
2
https://www.ictstandard.org/
334
Table 1. Hits
Database
ACM
IEEE
Elsevier
Wiley
EBSCOhost
Total
Title:
ITG
9
103
28
49
23
212
AND
Abstract:
ITM
2
4
2
15
1
24
AND
Abstract:
SM
3
4
7
AND
Abstract:
EA
2
1
3
Table 2. Appraised hits and selected papers

Database
ACM
IEEE
Elsevier
Wiley
EBSCOhost
Total
Available in full version,

written in English, scientific conference or journal
paper
2
8
2
7
1
20
Selected
papers
1
7
1
2
11
We selected 11 papers [11-21] which included a

statement for IT governance and the abstract
contained the term IT management, service
management or enterprise architecture.
There is one framework [21] where IT governance contains enterprise architecture and risk
management, which are connected with program management, IT investment management
and standards-policies-procedures by alignment
of strategies, processes and applications. Furthermore, there is one hierarchical structure of
IT governance [16] where people, goal, process
and technology are domains; strategy and tactics are scopes; understanding, decide and
monitor are decision-making phases. The third
finding is the conceptual view of the model for
IT governance performance prediction where IT
processes are evaluated by maturity indicators
because it is reasonable to believe that some of
the IT governance maturity indicators are correlated with IT governance performance [11].
Furthermore, performance is mentioned in the
IT governance statements [12, 15, 20] as follows:
IT Governance is defined to be a subset
discipline of Corporate Governance
focused on information technology (IT)
systems and their performance and risk

management. [12]
IT governance is the process by which
decisions
are
made
around
IT
investments.
A
well-matured
IT
governance framework is based on three
major elements: structure, process and
communication. In addition, there are four
objectives that drives IT governance and
must be covered in IT governance
processes: IT value and alignment,
accountability, performance measurement,
and risk management. [15]
IT governance includes the roles and
responsibilities used to apply information
systems and related technologies to
manage and support the organizations
functions. These roles and responsibilities
of IT governance fall into three domains:
the IT department, the users, and
management.
While
management
oversees the IT department, ensuring that
IT objectives link to organizational
objectives, users are the ones who monitor
IT systems and provide input to new IT
implementation plans. IT governance
objectives should emphasize:
1. Alignment of IT objectives to the
overall business strategy,
2. Measures of IT performance, and
3. Competitive advantages provided
by IT for the organization. [20]
335
3 SURVEY
The online survey was attended by five municipalities in the Northern Savonia, the two
federations of municipalities and one ICT
service provider. There were 331 potential
respondents and we got 136 answers. The
response rate was 41,1 % and it is excellent
because we did not have earlier connection to
the respondents (a response rate 30 % is acceptable in online surveys8). Our survey has
25 questions based mainly on the ICT Standard for Management [6]. However, in this
paper we report the answers of five questions
(Table 3) that position the ICT governance in
the Northern Savonia.
ICT governance has to work closely with organizations
business,
making
those
work products, it has agreed upon with business [6]. ICT can be good way to increase the
cost-effectiveness of business and in the best
ICT scenario can even offer totally new business possibilities. Therefore, we were interested how well this development collaboration works in responding organizations (the
question Collaboration works with business
and ICT?).
ICT strategy is an important tool for defining
the right direction for ICT governance - making sure that ICT governance follows the
overall strategy of the organization [6]. Models of ICT governance define responsibilities,
duties, objectives and how things of the different kind should be measured [6]. We were
interested in how respondents view the importance of the strategy (the question Need
for ICT strategy?) and the ICT governance
model (the question Need for ICT governance model?) within their organization.
In development projects it is normally the
duty of business to define need for development, but ICT governance usually documents
these need. ICT governance should also help
in evaluation between different development
options [6]. We were interested about how

early ICT is normally consulted in development projects (the question ICT is consulted
before development projects?).
Enterprise architecture helps making sure that
information systems work properly and support work well, as well as managing to risk
and costs of systems. It is the tool especially
for managing development as a whole. In the
survey, we were interested about how well
known the concept of enterprise architecture
was within attending organizations (the question Familiar with enterprise architecture?)
- so that current situation could be estimated
and improved.
Table 3. Sums of YES and NO answers
Questions versus sum
of answers
SUM of NO
answers
Collaboration
works
with business and ICT?
ICT is consulted before
development projects?
Need for ICT strategy?
Need for ICT governance model?
Familiar with enterprise architecture?
SUM
YES
swers
80
56
48
88
48
88
20
116
87
49
of
an-
We analyzed the results with JMP 119 that

allows us to illustrate differences between the
answers of the questions. We compared the
means between two or more correlated variables (i.e. SUM of YES answers, SUM of NO
answers) and assessed the differences.
First, we compared the answer groups around
the question Collaboration works with business and ICT? (Table 4), and we realized that
there is the biggest assessed difference with
the question Need for ICT strategy? (Figure
1). Furthermore, we realized that there is one
difference where the mean mean of NO answers is bidder that the mean mean of YES
answers (Figure 2).
https://www.utexas.edu/academic/ctl/assessment/iar/teaching/gather/
method/survey-Response.php
9
http://www.jmp.com/software/jmp/
336
Table 4. Answer groups around the question

Collaboration works with business and ICT?
Collaboration
works
with business and ICT?
versus
Familiar with enterprise
architecture?
Mean Mean
of NO answers
Mean
Mean
YES
swers
64
72
64
72
50
86
83,5
52,5
of
an-
Figure 2. Example of the assessed differences around

the question Collaboration works with business and
ICT?
Second, we compared the answer groups

around the question ICT is consulted before
development projects? (Table 5), and we realized that there is the biggest assessed difference with the question Need for ICT strategy?.
Table 5. Answer groups around the question ICT is
consulted before development projects?
the question Collaboration works with business and
ICT?

versus Question
Mean Mean
of NO answers

architecture?
48
Mean
Mean
of
YES
answers
88
34
102
67,5
68,5
337
Third, we compared the answer groups

around the question Need for ICT strategy?
(Table 6), and we realized that there is the biggest assessed difference with the question
Need for ICT governance? (Figure 3). There
does not seems to be differences between answers group with the question Familiar with
enterprise architecture?. However, there
seems to be differences between the answers
group with the questions Need for ICT governance model? and Familiar with enterprise architecture? (Figure 4).

versus Question

architecture?
Mean Mean
of NO answers
Mean
Mean
YES
swers
34
102
67,5
68,5
of
an-

the question Need for ICT strategy?
Fourth, we compared the answer groups

around the question ICT is consulted before
development projects? (Table 7), and we realized that there is the biggest assessed difference with the question Need for ICT strategy?.

Need for ICT governance model? versus
Question
Mean Mean
of NO answers

architecture?
53,5
Mean
Mean
YES
swers
of
an-
82,5

the question Need for ICT strategy?
338
4 DISCUSSIONS
There is a need for a simple framework supporting enterprise architecture and ICT governance. The framework should enable dialogue between various stakeholders. Furthermore, it is needed to ensure consistency in
decision-making. In the conceptual context
(Figure 5), we will combine key stakeholders
(users, customers, and providers of the different kind) and two main principles of governance frameworks (conformance and performance) with the entities around the service,
we will see that conformance will be established when the service meets the requirements which are specified in the contracts.
Furthermore, performance is deemed to be the
fulfillment of the obligations, which are based
on contracts or objectives.
Assessment of the enterprise architecture

work and ICT governance will be useful when
developing of organization and sequencing
and identification of milestones needed. Primary objective of our survey was to gather
information about the current status of things
of the enterprise architecture and ICT governance. Secondary objective was to raise interest and awareness to the things of the enterprise architecture and ICT governance discreetly. In future, we are going to repeat the survey to assess the commitment the enterprise architecture work and ICT governance in the Northern
Savonia. Furthermore, we try to find out reasons
for not knowing of the things of the enterprise
architecture and ICT governance.
Figure 5. Conceptual context [22]
The services are realized through logical application components and implemented on
logical technology components in TOGAF
Content Metamodel. When we talk about entities, which are related to service, we have
identified at least four entities (contract, objective, measure, and service quality), which
should be taken into account when the responsibility for information management is
shared. The objective, as well as, the service,
is tracked against the measure. The service is
governed and measured by contract. Furthermore, the service should meet the service
quality.
339
REFERENCES
[13]
[1]
ISO/IEC, Consolidated ISO Supplement

Procedures specific to ISO, Directives, Part 1,
Fourth
edition,
2013,
http://isotc.iso.org/livelink/livelink/fetch/2000/212
2/4230450/4230452/ISO_IEC_Directives%2C_Pa
rt_1_and_Consolidated_ISO_Supplement__Procedures_specific_to_ISO%2C_4th_edition_2
013_%28PDF_format%29.pdf?nodeid=14883571
&vernum=-2
M. Sall, G. Di-Vitantonio, Business Service

Management: The Impact of IT Governance
Models on IT Management Policies, IEEE
International Conference on Services Computing
(SCC'O6), IEEE 2006, pp. 373-380.
[14]
S. Saetang, A. Haider, CIO and CTO nexus:

Empowering organizations with IT governance,
Proceedings of Technology Management for
Emerging Technologies PICMET 2012, pp. 36813691.
[15]
A. Nabiollahi, S. bin Sahibuddin, Considering

Service Strategy in ITIL V3 as a Framework for IT
Governance,
International
Symposium
on
Information Technology, ITSim 2008. pp. 1-6.
[2]
The Open Group, TOGAF - Content Metamodel,

TOGAF
9.1.
http://pubs.opengroup.org/architecture/togaf91doc/arch/chap34.html
[3]
Finnish Standards Association SFS, Corporate

Governance of Information Technology ISO/IEC
38500:2008.
[16]
S.H. Cho, S.H. Lee, Fuzzy Decision making of it

governance, Proceedings of the International
Conference on e-Business (ICE-B) 2010, pp. 1-5.
[4]
International standard ISO/IEC 38500:2008,

Corporate
governance
of
information
technology.ISO/IEC, Information technology.
Service management. Part 1: Service management
system requirements, International standard
ISO/IEC 20000-1:2011
[17]
N, Zarvc, C. Stolze, M. Boehm, O. Thomas,

Dependency-based IT Governance practices in
inter-organisational collaborations: A graph-driven
elaboration. International Journal of Information
Management, vol. 32, 2012, p. 541549.
[18]
[5]
L 10.6.2011/634. Laki julkisen hallinnon

tietohallinnon ohjauksesta. Oikeusministeri,
Finlex,
http://www.finlex.fi/fi/laki/ajantasa/2011/2011063
4
M.-S. Pang, IT governance and business value in

the public sector organizations - The role of
elected representatives in IT governance and its
impact on IT value in U.S. state governments,
Decision Support Systems, Volume 59, March
2014, pp. 274-285.
[6]
ICT Standard Forum. ICT Standard for

Management, 2012, ISBN 978-952-67383-2-1..
[19]
[7]
ICT
Standard
Forum,
Tietohallintomallin
soveltamisohje julkiselle hallinnolle, Laine Direct
Oy, 2013.
C. Becker, G. Antunes, J. Barateiro, R. Vieira, J.

Borbinha, Control Objectives for DP: Digital
Preservation as an Integrated Part of IT
Governance, Proceedings of the American Society
for Information Science and Technology, volume
48, Issue 1, 2011, pp. 110.
[20]
[8]
H. Suonuuti, Guide to Terminology. Helsinki:

Tekniikan sanastokeskus ry, 2001.
L.N. Higgins, D.T. Sinclair, A new look at IT

governance, Journal of Corporate Accounting &
Finance, volume 19, Issue 5, 2008, pp. 31-36.
[9]
C. Okoli, K. Schabram, A Guide to Conducting a

Systematic Literature Review of Information
Systems Research, Sprouts: Working Papers on
Information Systems, vol. 10, 2010.
[21]
J.R. Getter, Enterprise Architecture and IT

Governance: A Risk-Based Approach, 40th
Annual Hawaii International Conference on
System Sciences. HICSS 2007.
[10]
B. Kitchenham, Procedures for performing

systematic reviews. Keele University Technical
Report TR/SE-0401, ISSN:1353-7776, 2004.
[22]
Lahtela A., Hotti V., Salomaa H.: Service Support

in IT Governance, IT Management and Enterprise
Architecture Context. The Fourth International
[11]
M. Simonsson, R. Lagerstrm, P. Johnson, A

Bayesian Network for IT Governance Performance
Prediction, ICEC '08: Proceedings of the 10th
international conference on Electronic commerce,
ACM 2008.
[12]
D. Radovanovic, M. Sarac, S. Adamovic, D.

Lucic, Necessity of IT Service management and IT
Governance,.MIPRO, 2011 Proceedings of the
34th International Convention, 2011, p. 14301433. ISBN: 978-0-9891305-4-7 2014 SDIWC
171
340
Perspective for Suggestion about Communication Plan / Strategy of Thailand

Using for ASEAN One Community Relevant to the 3 Pillars: APSC, AEC, and
ASCC
Nattanun Siricharoen
Lecturer, Faculty of Communication of Arts, Huachiew Chalermprakiet University
PhD [Candidate] STOU: Sukhothai Thammathirat Open University
Email: nattanun2004@yahoo.com
Waralak V. Siricharoen
Lecturer, School of Science and Technology, University of the Thai Chamber of Commerce Email:
waralak_von@utcc.a.th
Abstract:
Keywords:
ASEAN leaders agreed to shorten the time to

establish and integration of the ASEAN
Community by year 2015; the community has
been divided into 3 different communities which
are ASEAN Political-Security Community-APSC,
ASEAN Economic Community-AEC, ASEAN
Socio-Cultural
Community-ASCC.
A
communication plan is significant in order to
achieve common understanding. With well
understood, ASEANs activities can be of
considerable advantage to every country.
Thailands government policy has focused on
preparing the country to become an ASEAN
member in 2015 through its national development
strategies to build and strengthen the economy,
social and cultural affairs, and international trade.
Strategy for Thailand in the ASEAN One
Community are Education, in this paper provides
the case studies of how Thai University prepare
the Thai students for year 2015 to become
ASEAN one community. Three Thai University
are mentioned in this paper are Sukhothai
Thammathirat
Open
University:
STOU,
University of the Thai Chamber of Commerce:
UTCC and Huachiew Chalermprakiet University:
HCU. This examples show that there are
approaches to teaching and learning to prepare
people for ASEAN by offering 4 main skills to
step into ASEAN and a future career in KCCM:
Knowledge, Communication Skill, Creativity, and
Morality.
Communication Plan, Communication Strategy,

Thailand, ASEAN One Community, the 3 Pillars,
ASEAN Political-Security Community-APSC,
ASEAN Economic Community-AEC, ASEAN
Socio-Cultural Community-ASCC
INTRODUCTION
At the 12th ASEAN Summit in Cebu,
Philippines, ASEAN leaders agreed to
shorten the time to establish and integration of
the ASEAN Community by five years, to
2015; the community has been divided into 3
different communities such that each
congregation has the following goals (Tanesh
Kongprasert, 2012).
1. ASEAN Political-Security CommunityAPSC aims to promote cooperation in
political and security affairs; to strengthen and
maintain the peace and stability of the region;
to make each country live together in
peace; and to resolve problems and conflicts
when possible by peaceful means.
2 ASEAN Economic Community-AEC also
aims to create ASEAN as a single market and
production base and to make the movement of
goods, services, investment, and skilled
labor free.
341
3. ASEAN Socio-Cultural CommunityASCC aims to enhance the quality of life of

the population in member countries, the wellbeing, a good environment, and a feeling of
unity for the deeper cooperation in six areas
(Acharapan Taiyarat, 2012).
1) Development of human resources
2) Protection and social welfare
3) Rights and social justice
4) Environmental sustainability
5) Creation of an ASEAN identity
6) Reduction of the development gap
Among ASEAN (ASEAN Community) is a
cooperative agreement with the most
progressive cooperative agreement and the
emergence of this community in 2015 will
have an impact and make an extensive
difference in the socio- economic aspects of
member
countries extensive. In the
establishment of the ASEAN Economic
Community (AEC) is a draft or blueprint
(Blueprint), which is an integrated plan for
economic performance to achieve the
following four goals:
1) To make ASEAN a single market and
production base. There will be a movement of
goods, services, investments, and skilled
workers and more liberal capital movements
to follow the country's original ASEAN 6
member countries (i.e., Thailand, Malaysia,
Indonesia, the Philippines, Singapore and
Brunei) in gradually reducing or eliminating
trade barriers to non-tax (non-tariff barriers)
and marketing services sectors with
investment liberalization by 2015.
2) Strengthen capacity in the economic
competitiveness of ASEAN with a focus on
policies that will promote economic
integration: competition policy, consumer
protection, intellectual property rights, tax
policy and infrastructure
development,
finance, transportation,
information
technology, and energy.
3) Economic development through equity by

promoting the development of small and
medium-sized enterprises (SMEs) to reduce
the
gap of economic development between
member countries.
4) Integration into the global economy by
coordinating ASEAN economic policy with
countries in other regions to achieve
economic cooperation, such as the preparation
area. FTA with ASEAN dialogue partners is
encouraged to create a network in the field.
Production and regional distribution linked
to the global economy, etc.
Operations to move goods freely have set up
guidelines, for example;
1. Member States will have to reduce tariffs
on more than 8,300 items of imported goods
to 0 percent with the exception of some
products that have been designated as
sensitive products (sensitive list) or very
sensitive products (highly sensitive list) to
protect local manufacturers in the country.
2. The liberalization of trade in services.
Member States will have to reduce the
barriers to entry into the market for all service
branches and all forms of service. The group
covers
12
sectors
or
professions:
communications, construction, distribution,
education, environment, finance, health,
transportation, tourism, recreation, computer,
telecommunications and other fields such a
beauty spa, etc. In 2010, equity participation
was allowed for the four branches of
computer technology, health, tourism, and
air transport by giving investors a stake of
more than 70 percent.
3. Liberalization of investment ASEAN.
ASEAN must treat investors as well as their
foreign investors. ASEAN Investment
Agreement, or the ASEAN Comprehensive
Investment Agreement, includes major
investments such as investment promotion,
342
Trade liberalization and facilitation to invest,

and
investment
protection
covering
manufacturing,
agriculture,
fisheries,
forestry, mining, and related services.
4. The liberalization of the capital AEC will
have the opportunity to have a combination
of
financial
and
capital markets to
facilitate trade, investment, and increased
capital flows in the region.
Communication Plan
A communication plan is important to manage
to achieve mutual understanding. If the people
of each country can communicate and be well
understood, ASEANs activities can be of
significant benefit to every country. Using all
types of media is necessary, but in this digital
age, we all need to recognize that digital
media is very important to build cooperation
in various fields quickly and efficiently.
Trend Digital Media and interesting statistical
information in Asia and Thailand
A team from the advertising agency
ZenithOptimedia (2012), lecturing on the
topic of Trends of Digital Media in Thailand,
found that there are Top 3 web sites that
Internet users use in Thailand. They are;
http://google.co.th
or
http://google.com,
Facebook, and YouTube. Thais used them
for information sharing and networking. Their
conlusions of how Asians use the Internet and
social media are shown in Infographic: AsiaPacific Social Media Statistics in Figures 1, 2,
3, and 4.
Figure 1. Infographic: Internet user growth (20002010)

Resource:
http://www.digitalbuzzblog.com/infographicasiapacific-social-media-statistics-stats-facts/
Figure 2. Infographic: Internet and Mobile Penetration

Resource:
In Figure 1 the most Internet user growth is

12,035% in 10 years which is Vietnam. China
has the second-highest user growth with
1,767%. India has the third-highest user
growth with 1,520%. Indonesia has the
fourth-highest user growth with 1,400%. The
fifth is the Philippines with 1,385%. Thailand
is the sixth with 660% user growth. In Figure
2 is the comparison between Internet
penetration and mobile penetration of
343
Australia and countries in Asia. The highest

Internet usage is in South Korea and
Australia, followed by Japan, Singapore,
Taiwan, and Hong Kong. The highest mobile
usage is in Singapore and Hong Kong;
followed by Thailand, Taiwan, Vietnam, and
Malaysia.
Figure 4. Social Networks Population in Thailand

2013.
Resource:
http://www.slideshare.net/imcinstitute/thailandictoverview-q2-2013
Figure 3. Infographic: Top Social Media Usage in Asia

Resource:
The latest information on Thailand ICT Q2

2013, compiled by IMC Institute; September
2013, mentioned that the Top 4 social media
that Thais use is as shown in Figure 4. The
most popular is Facebook, with 18.5 million
users, which can be compared with the
previous information, shown in Figure 3, that
in 2010 there were 10.6 million Facebook
users in Thailand. The second is Line
Application on mobile phones with 18 million
users. The third is Twitter with 2 million users
and the fourth is YouTube with 5.3 million
users.
Figure 5. Show 92.8% of people entering Thailand

through Google search by +1274% growth in mobile
access.
Resource:
http://storify.com/imsakulsri/digitalmedia-trend-inthailand
344
Figure 6. Shows that in Thailand 14.9 million people

use Facebook = 21.4% of the total population,
accounting for 81.3% of the people who use the
Internet. Of people accessing Facebook in
2012, users aged 45-64 increased the most.
Resource:
http://storify.com/imsakulsri/digitalmedia-trend-in-thailand
Figure 8: Trend growth of the Internet in Thailand,

most of Social media followed by entertainment / email
/
searching
mobile
banking.
Resource:
http://storify.com/imsakulsri/digital-media-trend-inthailand
Figure 9: Internet access of people aged 15-19 years

mostly from 18:00 to 21:00 am = Prime time digital
ads / online ads & promotion.
Resource: http://storify.com/imsakulsri/digital-mediatrend-in-thailand
Figure 7. 6.6% of Thai people buy products and goods

by shopping online. The site with the most is
weloveshopping.
Resource:
345
Figure 10.
Thai youths find educational
videos from YouTube most often.
Resource:
Offering a Digital Media Perspective

on ASEAN
A survey on digital media trends in Thailand
made it clear that users knowledge and skills
are very important in modern times. As in
every field, skills are necessary to learn, work,
do business, etc. In all political, security,
economic,
and
social and
cultural
activities digital media is used. Therefore, it is
important to teach all individuals digital
literacy to avoid the loss of money and the
physical and psychological pain caused by
peoples deception in using digital media.
Digital information and data can benefit
people in Asian countries and the media can
be used efficiently and cost effectively.
ICT for ASEAN One Community
Thailands government policy has focused
on preparing the country to become an
ASEAN member in 2015 through its
national development strategies to build and
strengthen the economy, social and cultural
affairs, and international trade. The
preparatory action plan by the Ministry of
Information and Communication Technology
(ICT) established six strategic areas: Strategy
1:
Economic
change
(economic
transformation) is needed to prepare an

appropriate environment for business in order
t o a t t ra c t t r a d e , i n v e s t m e n t , a n d
business creation in the field of information
technology and communications.
Strategy 2: The participation of the
people and capacity building for the
people (people empowerment and
engagement) will improve the quality
of life through easier access to ICT at
reasonable prices.
Strategy 3: Innovation will promote
the information and communication
technology industry about the
environment (going green) by using
creativity and novelty, including the
promotion of research and innovation
in striving for academic excellence.
Strategy
4: Infrastructure
development for ICT will support
services, information technology, and
communications
across
all
communities in the region, including
the management and integration of
meteorological information warning
systems
and effective and time
disaster management.
Strategy
5:
Human
capital
development will involve human
resources with ICT skills and
capabilities to support the growth of
the ICT industry and help promote
other industries.
Strategy 6: Reducing the digital gap
will develop and put ICT to good use
on a daily basis to build careers.
The action plan to achieve the ASEAN
Communitys ICT Ministry herein guidance
to agencies and governments and relevant
private was conducted within the ASEAN
Telecommunications
and
Information
Technology
conference
and
ASEAN
including coordination and oversight of
projects and activities. Information and
Communication Technology Master Plan of
ASEAN and under the framework of ASEAN
cooperation involved. The action plan will be
reviewed every year for updates and
346
additional projects/ activities. There may be a

need to set up to meet mission readiness
towards the ASEAN Community in Thailand
(Chaiyun Peangkeiatpairog, 2014).
Ms. Areewan Howerangsi (2013), a
consultant to the Foreign Ministry, said the
government and the Ministry of ICT have
made ICT a priority. The government has
established a national strategy that includes
the following:
Strategy 1: Create
a
competitive
advantage over other countries.
Strategy 2: Decrease the differences and
create equal opportunities in society.
Strategy 3: Build an environmentally
friendly infrastructure.
Strategy 4: Modify the administrati on
with good governance for a balanced
system administration. Use the strategic
national approach to deal with issues
related to the preparation to join the
ASEAN community.
The mission of the Ministry of ICT relates
to the three pillars of the ASEAN community:
1) services that move freely, such as computer
services and telecommunications services; 2)
the development of an information technology
infrastructure and e-commerce; and 3)
promote the use of ICT among disadvantaged
children and women, including the
management of disasters and the security of
information (cyber-security).
Strategy for Thailand in the ASEAN One
Community
Education for ASEAN One Community
In education, there will be an operating
mechanism that includes sectorial meetings
among senior officials and the Council of
Ministers and the ASEAN Socio-Cultural
Community to achieve ASEANs caring and
sharing goals. The Declaration Cha Am-Hua
Hin on strengthening cooperation in education
will be reviewed. The role of education in
building the ASEAN Community by 2015

will involve according three pillars:
1. Role of academia in the political and
security pillar
Support, u n d e r s t a n d i n g , a n d
a w a r e n e s s about the ASEAN
Charter; more ASEAN charter schools;
and publications in the local languages
of the nations in ASEAN to focus
even more on the principles of
democracy.
Respect for human rights and values
in ways that support peace in the
school curriculum.
Understanding and awareness of
cultural diversity, regional traditions,
and beliefs among teachers trained in
exchange programs.
Establishment
of
basic
information online so school leaders
can meeting regularly and exchange
views on various issues in the region.
Capacity
building
and
networking, including recognition of
the existence of the schools in the
Southeast Asia (Southeast
Asia
School Principals Forum: SEA-SPF)
2. Role of education in the economic pillar
Content development in ASEAN

for schools to serve as a reference point for
training and coaching teachers by offering a
degree program in universities and proposing
a national ASEAN language.
A foreign-language elective in school

to support regional projects focused on
promoting awareness among the youth about
ASEAN and the endorsement of other
initiatives, such as the ASEAN Schools Tour,
ASEAN Student Exchange Program, ASEAN
Youth Cultural Conference, and Youth
Summit among ASEAN university students.
The ASEAN University Network

and youth speech contest to support lifelong
learning in ASEAN countries.
347
A
meeting
of
ASEAN
to
promote research, education, research and
development, cooperation, and to provide a
forum for researchers from member countries
to exchange views on issues and matters
related regional support, understanding and
awareness of awareness of the issues and the
environment by integrating the curriculums in
schools.
Provide awards for green schools

during ASEAN Day celebrations and
consider the establishment of a regional
development fund for education to ensure that
participants
receive
adequate financial
support for their operations.
Allow the ASEAN Sectorial Ministry

and ASEAN to continue to follow up on this
mission by providing guidance and support
for a 5-year plan of ASEAN on the subject of
education.
On Monday, August 23, 2010, there was a
meeting at the Ministry of Education
involving a national committee to promote
education in the region in preparation to join
the ASEAN Community in 2015. The
meeting approved the draft policy and the
education policy, as well as the following five
policies:
Policy 1. The provision of educated

news information and about ASEAN to raise
awareness and prepare credentialed teachers
and educational personnel, students, and the
public to step into the ASEAN Community by
the year 2015.
Policy 2. The development of

students and the public with the right skills to
join ASEAN, such as learning English and a
neighboring
language
and
learning
information technology skills and expertise to
adapt to changes in line industries and
thereby increase the chances of finding jobs
for all people to increase production.
Policy 3. The development of policies

to promote educational standards for the
students and teachers in the region and the

recognition of academic qualifications as
ASEAN promotes cooperation
among
various institutions and youth exchanges for
distance education in support of an education
for a lifetime; the promotion and
improvement of
education,
vocational
education, and professional training upstream
and downstream and the promotion and
enhancement
of
cooperation
among
institutions of the member countries of
ASEAN.
Policy 4. The preparation of a policy

to liberalize education in ASEAN, mutual
recognition of educational talent development
experiences in various fields to support the
liberalization of education, coupled with the
liberalization of labor mobility.
Policy 5. The development of

youth missions as a key resource in advancing
the ASEAN community.
In promoting university education in
Thailand, various fields will need to adjust
and prepare as follows (Thai Post Newspaper,
2012):
1. Adaptation and preparedness of the
agency is the responsibility of the
Ministry of Education and other relevant
ministries. The agency should have a policy
on higher education institutions and
strategic plans in support of higher education
to prepare Thailand to step into ASEAN and
put this into practice in the various
universities. Both state and private sectors
are part of the value chain.
2. Adaptation in the preparation of teachers
and educational personnel in different
universities. The need to develop their own
potential for learning, teaching, research,
academic services, and the preservation of
culture, particularly in English, which is the
common language of communication in the
region. But the current problem is that some
348
universities employees, both academic and
support call status, face insecurity, and since
they lack a contract, ties to the systems
stability are insecure. They depend on the
evaluation system to gain a contract,
indicating instability in the profession. Also,
the management or governance of the
Executive Committee of the University,
which affects the status of teachers, has
changed. The potential and efficiency of
teaching and learning in Thailand and
increasing the quality of education have been
compromised to the extreme, and no system
to strengthen and encourage employees has
developed potential in both languages. And
further education to higher levels adequate to
support the ASEAN community by entering
related and affiliated university itself is
considered an important factor affecting
Thailands tertiary education and preparation
to step into the ASEAN community.
Governments and related agencies must take
measures to resolve the issue ASAP and
must have a clear policy on human resources,
as the need for this in such studies is
substantial.
3. Adaptation to prepare the student. It is
inevitable that students will have to learn to
adapt and prepare for the future, as students
must be interested in and aware of the
consequences stemming from the integration
of various countries. ASEAN, in terms of
advantages and disadvantages, is complex to
fully grasp. The emergence of the ASEAN
community, which is reflective of modern
society, is increasingly a multicultural society.
A new generation of students is so necessary
to adjust the learning process, as an attitude
adjustment that remains aware of nationhood
is vital. The modern paradigm of learning
should be a goal, including such aspects as
knowing the situation, creating the ability
to work with others of different cultures, and
learning about the history and culture of
neighboring countries to achieve a better

understanding between them. With the
opportunity to learn a foreign language and
to increase skills in the English language,
students can learn to communicate very well.
4. Adaptation and preparedness of local
communities to step into ASEAN. Because
if the local community does not support
measures or plans to step into ASEAN, it will
be a disadvantage in the competitive market.
In addition, the local culture is something that
is paired with a social Thailand and the
creation of an ASEAN economic community.
Local communities need a leader with a
vision who has the ability to link the
economic integration of ASEAN to the
global economy. The local institutions
must have support in the coordination and
implementation of public relations and
providing knowledge
about
ASEAN.
Advantages and disadvantages to going to
the local community in the future mean
that the community leaders and local
citizens must learn and adapt to the local
community in planning for the next step in the
ASEAN community.
Mr.Chaiyaphruek Serirak (2013), Secretary
of the Office of Vocational Education,
discusses strategies to prepare for the ASEAN
Economic Community. The preliminary plan
includes increasing the capability for
competition of students studying to become
vocationally skilled technicians.
Dividing
the students into two groups to drive
efficiency includes designating the first group
as the group of students that is potent enough
to go out into international competition.
The course will be taught entirely in
International English starting in the year
2013. This course, which is now offered to 28
of 32 provinces, aims to reach 150 provinces
in year
349
2013. Meanwhile, to prepare the second group

of students to support the ASEAN
community, three aspects are currently weak,
including the following:
1. Language skills for careers
2. Information technology skills (ICT) for
careers
3.
Skills
in working with people of
different cultures
In addition to the undergraduate program,
vocational institutions that are ready will
coordinate moving students into private
internships as a preamble to opening 28
branches such as electrical and electronics,
automotive parts, travel, etc. Students will be
exposed to 30 students per classroom.
Bachelors Degree Occupational
line
practice to be performed which makes
vocational become a major force in the
production of quality personnel to support the
move towards liberalization of the ASEAN
Economic Community.
David M. Kennedy and Bob Fox (2013) said
in their research that students also agreed that
technologies, for example, social networking
software and web-conferencing, can improve
their communication with other students and
teachers and thus assist their studies.
There are interesting facts from the three
universities in Thailand; they offer the
international program for Thai and foreign
students. They use the ICT in different ways
to enhance the performance of teachers
and students, as in the following case studies.
Case study 1: Sukhothai Thammathirat Open
University: STOU
STOU offers Designing4Learning (D4L) as a
course-design process that was developed
especially for student-centered learning
environments. Students will get the
experience of distance learning with a teacher

and friends as though they are sitting in the
classroom with the teacher and their
classmates, as shown in Figure 11.
Figure 11. D4L log-in page on STOU home page

Resource: http://d4lp.stou.ac.th/?GID=13955
Figure 12. D4L Learner page on STOU home page

Resource: http://d4lp.stou.ac.th/?GID=13955
Case study 2: University of the Thai Chamber

of Commerce: UTCC
International programs
represent
the
academic expansion of the University of the
Thai Chamber of Commerce, in keeping with
global business trends. As of May 2012,
UTCC has 17,805 undergraduates and 1,464
graduate students. The total number includes
international students
from
different
countries
worldwide including China,
Myanmar, Brazil, Cambodia, East Timor,
Bhutan,
Laos, the Netherlands, Nigeria,
Russia, Japan, Bangladesh, Nepal, Finland,
350
Italy, Singapore, Ethiopia, Taiwan, South

Africa, Congo, Mexico, and the Philippines.
In the highly competitive and rapidly
changing environment, the curriculums
are incessantly improved and developed;
furthermore, new curriculums such as
Logistics Management and Global MBA,
Master of Science in Business Informatics
are offered to meet the demands of the new
economy.
David M. Kennedy, Bob Fox (2013) also
emphasize that the overall relationship
between students and technology is
significant, ubiquitous and a part of the
everyday life of a student, whether they are
engaged in social or educational activities.
Conole et al. (2006) categorized the student
views of technology and their opportunities
for learning into eight components. They were
the following: pervasive and integrated,
personalized, social, interactive, changing
skills set, transferability, time, and changing
work patterns.
Now, UTCC engaged the new technology and
the new way of communication in class to
assist the students, as seen in Figure 13 and
14, such that the students used iPad and
iPhone to make presentations in the class.
Figure 13. Thai Undergraduate students used iPhone

and Smart Phone with Slide Shark Application to
present their work
Figure 14. Thai Undergraduate students used iPad and

tablets to review what they will present in front of the
class.
Case study 3: Huachiew Chalermprakiet

University: HCU
HCU is a private university which also offers
international programs. The program aims to
enhance their English language skills and
knowledge about China and Chinese culture
together for supporting business growth in
China in the future and to prepare to study
abroad at a higher level, and also to provide
the opportunity for international students to
study the science and culture of Thailand.
There are two international undergraduate
majors, which are the Faculty of Liberal Arts,
offering a major in English for Chinese
students, and the Faculty of Business
Administration, offering a Bachelor of
Business in Chinese or English.
According to Marketingupdate.co.za (2013),
lecturers must adapt to the changing
environment of their students by meeting
them where they are most comfortable in the
social media space, and, as education expert
Natasha Madhav says, Generally speaking,
institutions continue to remain too passive to
the changing needs of their student body.
Madhav is the head of the Program of
Information Technology at the Independent
Institute of Education, South Africas largest
351
and most-accredited private higher education

institution.
For HCU, the lecturers are using Social Media
to connect with colleagues in the same
department or in the same faculty or school
for help and collaboration. Also using
Facebook to connect to students within the
class group as see in Figure 15, by setting the
group as a secret group, will help in
saving the information securely within the
group.
to support the various activities. As Thailand

is a member of ASEAN, we need to
emphasize and focus on three approaches, as
enumerated in the following recommendation:
1. Digital media for all
Facebook and other forms of social

networking media have become part of most
peoples lives, and almost without exception
part of the lives of the youth.
They
communicate, socialize, learn and entertain
themselves through the internet and more
especially using social platforms to
connect to each other, notes Madhav at
Marketingupdate.co.za (2013).
1.1 Each person must find their knowledge

and skills in the use of digital media in order
to outsmart the technologies that are changing
rapidly. Seeking out their own knowledge by
using the mobile/smart phone or internet is
the most convenient and possible way for the
new generation.
1.2
Government agencies and the
private sector need to help by supporting
learning via easier and cheaper access to
digital media. There are many free course
materials available online, called open
courseware,
from
the University of
Cambridge website.
1.3
Each country should cooperate in the
exchange program to learn digital media
between the people of their country and the
people of neighboring ASEAN countries in
order to achieve learning together and to help
with sharing new knowledge with each other.
For example, there are many joint programs in
which students study in a university in
Thailand and travel to have a short research
period and final defense exam in Myanmar.
Conclusion and Recommendations
2. ICT for all
Dr. Wiriyah Ruechaipanit (2012), founder of

www.eduzones.com, said that there are
approaches to teaching and learning to
prepare people for ASEAN by offering 4
main skills to step into ASEAN and a future
career
in
KCCM:
Knowledge,
Communication Skill, Creativity, and
Morality. This paper shows that the most
important aspect is to plan the use of digital
media, ICT, and modified forms of education
2.1 Learning about information technologies

and communications that change regularly.
Affordable technologies are everywhere, as
PC tablet smart phones are getting cheaper
and have better quality each year. The easy
interface and familiar menus with the touch
screen function help develop technology skills
and make the world smaller, bringing people
closer.
Figure 15: The Facebook page of a lecture group for

collaboration with colleagues and students
Resource: https://www.facebook.com/groups/
352
2.2 With funding from the government to

promote ICT training for entrepreneurs and
the general public. Small and medium
enterprises can start a business in one hour
with a website, blog, and social media. There
are many college/university students and even
high school students who are business
owners, selling products on the internet.
2.3 Arrange TV programs or presentations
about ICT knowledge through media
channels, TV, and Internet media. YouTube
and Internet TV became the major
broadcasting sources of entertaining and
knowledge; the cost is almost zero, the user
can be the teacher by posting content, and
users can be students at the same time by
watching.
2.4
In order to welcome 21st-century
learning, educators must be prepared for an
academic shift, to include technology, and to
create networked learning environments that
harness collaboration, learner participation,
problem solving, and the sharing of ideas.
Social networking sites can assist teachers and
students in collaborating.
3. Education for all
Integration of these digital media, ICT, and
education will lead to creating effective
communication plans or strategies for
Thailand, understanding each other, and
helping
each other, along with developing
ourselves and using for ASEAN One
Community the three relevant pillars: APSC,
AEC, and ASSC.
REFERENCES
[1] Acharapan Taiyarat (2012) Technology and
Innovation
in
Education.
Retrieved
from
http://atcharapun262.blogspot.com/2012_09_
01_archive.html
3.1
Public and private support for
scholarships for masters degrees and the PhD
in Management Education about digital
medias influence over ICT and with respect
to the cooperation of ASEAN in particular,
for example, ASEAN foundations with
Microsoft Youthspark help arrange website
creation for starting business training for
youth
inside
and
outside
the
school/university.
3.2 Promoting distance learning courses
between ASEAN countries to save time and
cost
while
preserving
knowledge,
understanding and beneficial cooperation
between ASEAN countries.
Sukhothai
Thammathirat Open University, STOU,
offers distance learning (e-learning) in
Undergraduate, Masters, and PhD levels
with their D4L software embedded in the
STOU website in order to assist students
and teachers in collaboration while traveling.
3.3
Promoting the use of English as the
second and third language in their own
countries to help the workers communicate
with each other and foreigners who come to
work in outsourcing companies in order to
support businesses communication.
[2] Areewan Howerangsi ( 2013) The action
continues their ICT knowledge. Preparing ASEAN.
Retrieved from http://www.mict.go.th/view/
[3] Chaiyaphruek Serirak (2013) Thailand to
boost
vocational Bachelor's Degree Career practical and
AEC.
Retrieved
from
http://www.dlfeschool.in.th/TETA/20130303 t.html
[4] ChaiYun Peangkeiatpairog (2014) Ministry of ICT
Action plan aimed to prepare Thailand for the Asean
Community.
Retrieved
from
http://www.thaigov.go.th/th/news-ministry
[5] Conole, G., de Laat, M., Dillon, T. & Darby, J.
(2006) JISC LXP: Student experiences of technologies
(Final Draft Report). Joint Information Systems
Committee
(JISC),
UK.
Retrieved
from:
http://labspace.open.ac.uk/file.php/1/kmap/1
176712833/references/LXP%20project%20fi
nal%20report%20dec%2006.pdf
353
[6] David M. Kennedy, Bob Fox (2013) Digital
natives: An Asian perspective for using learning
technologies. International Journal of Education and
Development using Information and Communication
Technology (IJEDICT), 2013, Vol. 9, Issue 1, pp. 6479.
[7] IMC Institute (2013) Information on Thailand ICT
Q2
2013.
Retrieved
from:
http://www.slideshare.net/imcinstitute/thailan
d-ictoverview-q2-2013.
[8] Marketingupdate.co.za (2013) Connecting the
lecturer to students 24/7 via route social media.
Retrieved
from
http://www.marketingupdate.co.za/?IDStory
=58318
[9] Tanesh Kongprasert (2012) Entering the
ASEAN
Economic Community (AEC) and its impact on
Thailand's
economy.
Retrieved
from
http://www.thaiworld.org/th/include/answer_
search.php?question_id=1174
[10] Thai Post Newspaper (2012) Education of
Thailand in the ASEAN context. Retrieved From
http://www.interscholarship.com/4545/9319
[11] Viriya Ruchaipanit (2012) Four key skills survival
to
ASEAN.
Retrieved
from
https://www.youtube.com/watch?v=o6k5LTy IqDM
[12] ZenithOptimedia and Performics (2012) Trend
Digital Media
in
Thailand.
Retrieved
from
http://storify.com/imsakulsri/digital-mediatrend-inthailand
354
An Instructional Design Model and Criteria for Designing and Developing Online
Virtual Labs
Mohamed Elsayed Ahmed 1, 3 and Shinobu Hasegawa 2
1
School of Information Science, Japan Advanced Institute of Science and Technology
2
Center for Graduate Education Initiative, Japan Advanced Institute of Science and Technology
3
Department of Educational technology, Faculty of Specific Education, South Valley University
1, 2
1-1, Asahidai, Nomi, Ishikawa, Japan, 3 83523 Qena, Egypt
1, 3
mohamedelsayed@jaist.ac.jp, 2 hasegawa@jaist.ac.jp
ABSTRACT
The purpose of this work is to propose a general
instructional design model to teach students in faculty
of education, especially in department of educational
technology how to design and develop online virtual
labs in a common way. We have made analyses of
previous instructional design models and related
studies in regard to the virtual labs to specify diverse
features of the proposed model. It was found that the
online virtual labs have no conventional instructional
design model, especially for designing and developing
stages and also no common shape and components.
Based on these results, we have reached to a new
suggestion model which guides the students in
refinement of the future learning environment using
recent technology. In this paper, we also present a list
of criteria for designing and developing the online
virtual labs.as a modern principles for directing
designing process of online virtual labs environments
to become instructional products. We have made a
derivation of these criteria from previous studies
related to the virtual labs, e-Learning technologies and
some miscellaneous technological resources in
educational technology. These criteria would provide
the students with educational and technological
guidelines to produce the online virtual labs with high
quality and efficiency.
KEYWORDS
Online Virtual Labs, Instructional Design Model,
Educational Technology Students, Designing and
Developing Criteria.
1. INTRODUCTION
In the 21st century, technological revolution had
made a great positive impact on diverse fields
including education. Therefore, ICT became one
of the main elements of any educational
systems/curriculums nowadays. This impact led
to the emergence of the many applications of
technology in education, such as e-learning,
distance learning, virtual learning, and virtual
campus/classroom. Some educational systems
have expanded in using e-learning at current

schools and universities because e-learning has a
potential for overcoming individual differences
among learners and for providing them with
learning environments anytime and anywhere.
Virtual labs (VLs) are one of the important
learning applications in developing modern
education. The VLs work as an alternative
solution of real labs since the VLs have few
restrictions about equipment, cost, and risks of
experiments. In addition, the VLs provide the
learners with sufficient training opportunity for
the experiments because of time and economic
efficiency. In recent studies, the VLs have gained
considerable attention in improving education,
especially in supporting practical learning. Chu
[1] claimed that the VLs provided simulations of
complex scientific processes that were less likely
to be demonstrated in the real lab. Furthermore,
Tatli and Ayas [2] confirmed the VLs enriched
the learners experiences as a supportive factor to
the real labs. On the other hand, many previous
studies have proven the important impact of the
VLs on learning practical skills in various fields
such as chemistry [3], computer network [4],
medicine [5], mathematics [6], and so on.
From these previous researches, the VLs lead to
mastery and creativity in learning by providing
interactive virtual environments, which make the
learners repeat diverse activities for improving
their skills and for conducting their experiments
with saving costs and full control in experiment
variables. In addition, we can now use the VLs for
theoretical fields that have no labs in the reality
by experiencing a new simulation to understand
difficult concepts and relationships. If we intend
to teach geography of climate and weather
practically, for example, we can use the VLs
including a simulation in regard to changes in
temperature on the map.
In order to improve modern education by using
such new technology, who should be responsible
persons? One of the candidates would be students
355
in faculty of education, especially in department

of educational technology. Because these students
should have potentials for implementing such
technology in the future schools and universities.
Besides, their academic programs in their
universities usually contain diverse ICT courses
which show them how to integrate modern
technology in coming education. Consequently,
such students have to develop their skills in
designing and developing the online virtual labs,
we call it DDOVLs, for the era of e-learning and
virtual learning. However, now they may have no
instructional design models which guide how to
design and develop the online VLs because most
of the previous studies about the VLs focused to
establish the effectiveness of the VLs in specific
fields as new tools in education.
In this paper, we suggest an instructional design
model to teach DDOVLs to the students in the
department of educational technology and in the
faculty of education through uniform shape and
components. This model would enable them to
reach suitable skills for DDOVLs at any courses
and fields in the future schools and universities.
In addition, we present a list of criteria to provide
guidelines of DDOVLs for the students as an
initial teaching tools for the proposed model. We
have made a derivation of the criteria, which
consist of objectives, contents, learning activities,
evaluation, feedback, platform, multimedia
elements, simulation, administration, navigation,
interactivity, publishing, and technical support
from literature reviews of the recent previous
studies in regard to the VLs and e-learning,
previous criteria of designing online courses, and
websites.
This paper is ordered as follows. First of all, we
describe target students as next responsible
persons for our DDOVLs model. And then, we
discuss a general definition of the VLs and the
online VLs without domain restrictions in section
3. After that, we claim support factors and
advantages of the OVLs in section 4 and 5. In
section 6, we describe required tasks for teachers
and learners in using the OVLs in the classroom.
In section 7 and 8, we propose an instructional
design model and suggest a DDOVLs criteria to
teach DDOVLs. We finally conclude our paper
and explain future work in section 9.
2. TARGET STUDENTS
Online virtual labs (OVLs) are new web-based
technology in improving modern practical

education. Therefore, most teachers and
instructional designers in the current school may
grope for a way to integrate such technology into
daily teaching and learning. This makes the main
targets of our research two types of the students
as the next generation of specialists and
responsible persons for ICT in education.
2.1 Students in Department of Educational
Technology
The students in the department of educational
technology typically learn how to integrate ICT in
education, according to educational theories,
instructional design, and other perceptions of
educational sciences. The range of their learning
usually covers designing, developing, utilizing,
and evaluating technological resources, such as ecourses, instructional websites, interactive videos,
digital images and graphics, and educational
software. Therefore, these students have
potentials to be responsible for implementing
such technology in the future schools after
graduation as instructional designers, content
developers, ICT teachers, and specialists for
educational technology.
2.2 Students in Faculty of Education
Most of the students in the faculty of education
will become teachers who specialize in one
subject like chemistry, physics, mathematics,
geography, English and so on. The range of their
learning mainly covers traditional pedagogy with
their specific fields. However, they are also
requested to understand the advantage about
implementing ICT in education as next generation
teachers. Therefore, these students will take an
important position for utilizing such technology
to the future classroom.
3. GENERAL VLs
3.1 General Definition of VLs
In an effort to teach the VLs to the students in the
faculty of education, we first need to consider
about a general definition of the VLs because they
should learn basic concept of the VLs without any
domain restrictions.
The definitions of the VLs in some literatures
have been limited as a substitute for the real labs.
Babateen [7] stated the VL is a virtual studying
and learning environment that stimulates the real
356
lab. It provides the students with tools, materials

and lab sets on computer so as to perform
experiments subjectively or within a group at
anywhere and anytime. Harms [8] also defined
the VL as a computer simulation which enables
essential functions of laboratory experiments to
be carried out on a computer. In addition,
Carnevali and Buttazzo [9] referred to the VL as
a computing system that allows to share the
physical resources available in a laboratory with
remote users connected on the internet. On the
other hand, it was clear the difference of the VLs
form in the previous studies. Harms [8] focused
on a computer simulation, Carnevali and
Buttazzo [9] represented a computing system,
and Prieto-Blzquez et al. [10] referred to virtual
learning environment.
Based on the above discussion, we define VLs are
virtual learning and teaching environments
designed and developed by means of the
instructional design models. In the VLs, the
learners interact with diverse kinds of learning
methods such as excises, practices, activities and
simulations to reach the learners goals such as
mastery of a specific skills and understanding of
relationships among concepts anytime and
anywhere.
3.2 General Design of Online VLs
In this section, we discuss about the general
elements of the VLs such as their shape and
components to reach to one uniform of the online
VLs (OVLs). The OVLs become now one of new
tools in developing education to enhance online
courses in any schools and universities. Therefore,
such uniform would be suitable for teaching the
virtual points in the OVLs to the students.
In order to consider the uniform of the OVLs, we
have made an analysis of many fields of the
previous studies which proposed not only online
but also computer-based VLs such as science [11],
chemistry [3], [12], biology [13], physics [14],
medicine [5], psychology [15], mathematics [6],
calculus [16], computer science [17], automatics
and robotics [18], computer network [4], [19],
software engineering [20], computer interface
technology [21], linguistic [22], studio [23],
history and cultural dynamics [24], statistics [25]
and others. After analyzing these previous studies,
we summarized the following results:
1. The significant impact of the VLs technology
was to increase learning skills and educational
2.
3.
4.
5.
attainment of the learners in the various fields.

The VLs are suitable for teaching courses
which have the real labs and have no lab in the
reality.
There was no fixed form in designing of the
VLs.
Components and way of organizing contents in
the VLs differed on each previous study.
The VLs were divided into two types, OVLs
and computer-based VLs.
Here, we first make clear the differences between

the two types of the VLs as follows.
OVLs: This type of the VLs provides virtual
environments in the form of the website via the
internet. Therefore, the learners can learn skills,
experiments and theoretical ideas about the
courses practically anytime and anywhere. In
addition, they interact with not only the contents
through interactive simulation, but also the
teacher and their colleagues. Such OVLs also
provide them with many information offered by
external website links.
Computer Based VLs: This type of the VLs is
available in the form of specialized software
which contains menus and tools in a specific field.
After setting up the VL program on the computer,
the learners can conduct experiments to obtain
practical skills. This type sometimes needs a lot
of time in designing and producing the program
package and is now limited to specific practical
fields, for example, ChemLab [26], Crocodile
Physics [27], and Virtual Chemistry Lab Program
[28].
From the results of the analysis, we decide to
suggest a unique and general form of the OVLs
suitable to any courses/domain (not only practical
but also theoretical) in educational institutions.
Our suggestion is divided into the following two
parts:
Fixed part: We call it OVL platform. This part is
stable in any OVLs. It consists of common
functions such as basic user interfaces to display
any virtual experiments from any courses,
management tools like login, tracking, and
communications tools such as e-mail, forum, chat,
and video conference.
Variable part: We call it OVL content. This part
is variable according to the contents of the courses.
In this part, we suggest how to organize the
contents inside the OVLs to become segments.
357
Each segment should belong to one experiment,

which consists of educational objectives,
theoretical
explanation,
instructions
and
procedures, simulation, activity, evaluation, and
summary.
Segment
Segment
Segment
Segment
Such form of the general OVLs would help the

students for designing and developing the high
quality OVLs in a stable way.
General OVLs
OVL Content
OVL Platform
Figure 1. Suggestion form of general OVLs
4. OVLs SUPPORT FACTORS

4.1 OVLs as Supportive Tools for Online
Courses
The OVLs and online courses are similar in
presenting something to the learners via the
internet in an easy and interactive way. However,
the types of these contents are quite different. The
online courses typically provide e-contents with
multimedia. On the other hand, the OVLs mainly
serve interactive simulations for practical
experiments. It means the OVLs would enhance
the online courses from practical point of view. In
order to expand the availability and reusability,
we sometimes implement both materials
independently. We can also integrate the OVLs
environment into the online courses as
complementary learning resources.
learning on the OVLs, they have to pass the

following several steps in order to learn the
contents deeply.
Firstly, the learners should identify general and
specific objectives in the OVLs and know the
theoretical background about simulations through
briefly text information and instructions about
how to perform the practical simulation in the
OVLs without mistakes. And then, they begin to
conduct simulation practically through many
actions such as dragging and dropping items,
changing values and shapes, collecting, arranging,
and classifying items. They may also
communicate with other learners with diverse
synchronous and/or asynchronous tools to
confirm their achievement and skills. Finally,
they have to make a self-evaluation through
answering electronic questions and read a
summary about main information of the target
simulations to conclude the contents.
The following figure shows these cyclic steps of
the idealistic learning process inside the OVLs. If
the learners do not acquire the target skills, they
can repeat learning from the Objective stage. As
a result of the learning process, they would reach
to the following outcomes:
1. Sufficient training about practical skills.
2. Learners experience related to the real labs
equipment.
3. Learners experience related to the virtual
learning environment tools.
4. Mastery of learning by training and
experiencing difficult theories practically.
5. Implementation of new ideas during learning.
4.2 OVLs as Enhancement Tools for Learning

inside Real Labs
Another role of the OVLs is to increase levels of
learning in the real labs. We previously claimed
that the OVLs are sometimes used in blended
learning where the learners can make training
before and after experience in a real lab [29]. On
the other hand, they may be able to learn new
simulation which has no real lab. In both ways of
Figure 2. Idealistic Learning process inside OVLs
358
5. ADVANTAGES OF OVLs
The OVLs are characterized by many advantages
as modern methods in developing educational
process. Some previous studies presented diverse
advantages of the VLs such as [3], [30], and [31].
Now we have organized and added these
advantages from improvement of learning point
of view as follows:
1. Economic: The OVLs are low in costs of
equipment and tools because of the virtual
form as well as of maintenance compared with
the real labs.
2. Richness: The OVLs enable to have rich
environments with the kits and tools,
furthermore with information through linking
to especial websites.
3. Safety: The OVLs provide safe experimental
environments, especially with regard to the
danger and harmful experiments.
4. Sufficiency: The OVLs supply sufficient and
repeatable
training
environments
for
experiments like learning by doing, which

lead to mastery in learning.
5. Consolidation: The OVLs help the learners in
improving their performance in terms of
experiments, equipment, and tools in the real
labs.
6. Motivation: The OVLs cause excitement in
learning to the learners by interactive
simulation about reality and theoretical ideas,
which encourage to continue their learning
process.
7. Understanding: The OVLs facilitate the
learners to reach a high level of understanding
especially in practical experiments, concepts,
laws, rules, relationships, processes, and ideas
in various courses, which leads to improve
learning.
8. Interactivity: The OVLs serve interactive
environments to the learners with not only the
OVL contents, but also the teachers and other
learners.
9. Accessibility: The OVLs provide diverse
contents to the learners anytime and anywhere.
10. Creation: The OVLs offer possibility of
original experiments by testing new variables
on the virtual representation about some
theoretical ideas, which conduce to creativity
in learning.
11. Integration: The OVLs work to enhance the

online courses through offering practical part
of the course.
6. TASKS OF TEACHER AND LEARNER IN
OVLs
6.1 Tasks of Teacher in Designing Process
In actual education systems, the teachers would
have many tasks in the designing process of the
OVLs as the content experts. In this section, we
illustrate two types of the tasks by the teachers
responsibility. Some educational systems focus to
qualify the teachers to specialize in one subject.
In addition, these systems also produce specialists
in educational technology to work together with
the teachers in implementing technology in the
schools. In these cases, the teachers play an
important role in designing the OVLs regarding to
their teaching courses. Therefore, they must be
familiar with teaching strategies for e-learning
and the technological ability. The main point of
their role is to participate with specialists of
educational technology in the following design
steps:
1. They must decide target learners and courses.
2. They must specify teaching topics from the
courses.
3. They must determine the general and specific
objectives and formulate the objectives
accurately.
4. They must identify the importance and
necessity of the OVLs for the target.
5. They must prepare the appropriate contents
and activities for the OVLs to achieve these
objectives.
6. They should consider how to organize the
topics logically.
7. They should select the suitable evaluation style
for each topic.
The teachers in other educational systems may
also have additional tasks in designing process to
integrate ICT into the classroom in the absence of
the specialists of educational technology in their
schools. In these cases, they have to be
responsible for not only designing and but also
developing the OVLs besides teaching courses.
We summarize such additional tasks as follows:
1. They must study and apply criteria for
designing and developing the OVLs with high
quality.
359
2. They must design and implement an OVL

platform.
3. They must design and implement OVL
contents.
4. They must publish the OVLs via the internet
and conduct technical tests.
5. They should hold training to other teachers
about how to use the OVLs.
6. They should conduct improvement of the
OVLs in light of the evaluation results.
6.2 Tasks of Teacher in Teaching with OVLs
When using the OVLs in teaching, the teachers
also face several tasks as follows:
1. They must plan online teaching to reach best
practice.
2. They should encourage and motivate selflearning for the learners.
3. They should make training to the learners
about how to use the OVLs.
4. They should plan to integrate the OVLs in the
real labs.
5. They should interact with the learners via the
internet in a synchronous or asynchronous
manner.
6. They should evaluate and monitor the
performance of the learners in order to
continue for improvement of education.
7. They should provide enrichment activities
with additional internet sites to get related
information easily.
6.3 Tasks of Learner in OVLs
In using the OVLs in learning, the learners are
required to pass the following tasks:
1. They must perform self-learning for the OVL
contents according to their intentions and
attitudes.
2. They must conduct the requested activities and
assignments and send them to the teachers via
electronic form.
3. They should contact with the teachers and
other learners through e-learning technology
such as e-mail, forums, chat, audio, and video.
4. They should expand creative thinking during
their learning process to reach new
relationships among the variables.
7. INSTRUCTIONAL DESIGN MODEL FOR

DDOVLs
In order to develop the next generation teachers
discussed in the above section, we suggest an
instructional design model for DDOVLs to teach
the students in the department of educational
technology and the faculty of education how to
design and develop the general OVLs suitable to
any field. The model would guide them to good
practitioners of the OVLs to improve education.
In order to reach to our suggestion model, we have
made an analysis of diverse previous studies. Some
of them confirmed the effectiveness of the proposed
OVLs [5], [6], [11], [12], [13], [14], [15], [17], [18].
Others discussed the previous models of the
instructional design such as ASSURE model [32],
ADDIE model [33], Kemp model [34], Dik and
Carey model [35], Stephen & Staley model [36],
Gerlach and Ely model [37]. The results of the
analysis are summarized as follows:
The OVLs had no instructional design model
specific in teaching how to design and develop
them for the teachers/specialists in developing
education, especially the students in the
faculty of education.
Most previous models focused on the
components of the phases of the traditional and
computer-aided education. For example,
educational software was not mentioned as the elearning and virtual learning components, which
now becomes an essential element of modern
educational systems.
A number of the previous models involved the
following phases: analysis, design, development,
and evaluation.
Consequently, we have decided to propose our
original instructional design model for DDOVLs as
described in Figure 3 because we believe it is
necessary to become widespread of teaching
DDOVLs in various educational institutions.
360
Figure 3. Instructional design model for DDOVLs
7.1 Pre-Design Phase

Pre-design is the first phase of the DDOVLs
model where the students specify the target
learners by determining the educational stage and
grade like 3rd grade of elementary schools or
freshman year of universities. Then, they select
a target course and specify course topics that they
intend to teach in the OVLs. Typically, they
would pick up the topics regarding to not only
recurrent training but also difficulty in
understanding. Finally, they decide general
objectives of the OVLs.
7.2 Criteria for DDOVLs Phase
The criteria are usually used to provide the
guidelines for the students when designing and
developing any technological tools for education.
The students should learn the criteria to reach to

acceptable educational and technological OVLs.
Therefore, the professors of educational
technology can prepare a list of criteria of
DDOVLs or select any lists of the criteria
prepared previously and teach it to the students.
7.3 Design Phase
The main role of the design phase is to provide the
students with the sequential steps in order to reach
general design suitable to any courses. As
described in Section 3.2, we propose in this paper
a new shape of the OVLs by dividing into the
following two parts.
361
7.3.1 Designing OVL Platform

As we described the above, the OVL platform is
the fixed and essential part in any OVLs. This part
contains a group of tools which provides the
learners with high interactivity during learning
inside the OVLs. It also supports the teacher to
manage learning and contents. It consists of the
following elements:
a) Determine Learning Management Tools
In this step, the students specify some tools to
manage learning inside the OVLs. The following
tools are typically used in the OVLs:
A registration tool which controls the learners
entrance in the OVLs.
A tracking tool which records the performance
of the learners in the OVLs.
A download tool which presents the learners
important files for learning inside the OVLs.
An announcement tool which provides the
learners with important dates for sending duties,
etc.
A search tool which supports the learners to find
additional information.
An evaluation tool which presents the learners a
couple of the types of evaluation.
b) Determine Communication Tools
Communication is one of the important parts in
learning on the OVLs because the learners are
able to obtain actual usage of the knowledge by
communicating with the teacher and other
learners. Thus, the students should determine
suitable communication tools. Some of them are
synchronized such as chat rooms or video/audio
conferencing. The others are asynchronous like
forums and e-mail.
c) Determine Tools for Displaying Contents
The contents are main components in the OVLs.
Thus, they should be presented easy to understand.
The students have to specify some tools to display
the OVLs contents such as upload tools, glossary
tools, and content update tools.
7.3.2 Designing OVL Contents
The OVL contents changeable according to the
courses or the fields. In this part, the students in
the department of educational technology need to
cooperate with some experts like actual teachers.
In the case of the students in the faculty of
education, however, all designing process would
be conducted by them. Consequently, we present

how to design the OVL contents for both students
through the following sequential steps:
a) Determine Behavioral Objectives:
The students divide the general objectives of the
OVLs into sub-objectives called behavioral
objectives. Each objective covers a certain topic
inside the OVLs to be easily observed and
measured.
b) Determine Actual Contents:
The students specify sufficient contents for each
topic that they intend to teach in the OVLs. They
also specify digital images and graphics, digital
audio, electronic drawing, and video so as to
achieve the learning objectives inside the OVLs
simply.
c) Organize Topics Logically:
The students arrange the list of the prepared topics
of the OVLs logically from easy to difficult or
simple to complex.
d) Determine Educational Activities:
The students specify suitable activities for each
topic in the OVLs. These activities enable the
learners to confirm information, increase their
motivation, and support critical thinking and
problem solving.
e) Determine Evaluation Style:
The students determine appropriate evaluation
style. The style should be suitable to each topic
with observance diversity in evaluation styles. In
addition, they also determine the evaluation style
for the whole of the OVL contents.
f) Design OVL Segments:
We propose how to design the OVL contents
through conversion of the OVL topics into
segments. Each topic belongs to one experiment
or other concepts, and should be matched by one
segment as described in figure 4. Each segment
should have a clear title and following tabs:
educational objectives, theoretical explanation,
instructions and procedures, simulation, activity,
evaluation, and summary as follows.
362
Figure 4. Organization of OVL contents
Educational Objectives: At first, the learners

recognize the educational objectives that they
should acquire at the end of learning for each
segment.
Theoretical Explanation: In this part, the OVLs
present the learners the theoretical backgrounds
for each segment such as scientific laws, rules,
and applications, information, and description.
Instructions and Procedures: This part explains
the correct sequences to conduct the experiment
in practice.
Simulation: It is a vital part as contained in the
virtual components for the experiment in the form
of images and animation, where provide the
learners with practical implementation by dealing
with several actions.
Activity: Each segment of the contents should
include educational activities. Such activities
enhance the learners performance and conduce
the improvement of learning outcomes and
mastery of skills.
Evaluation: Each segment must contain the
questions for self-assessment inside the OVLs.
Such questions would confirm mastery of the
contents and achievement of the objectives.
Summary: It must be placed at the end of each
segment. It makes the learners summarize the
main points in the OVL segment.
7.4 Development Phase
In this phase, the students begin to develop the
OVLs with application development
environments as the following steps:

a) Determine Computer Programs:
The students first select application development
environments to realize the designed OVLs with
high quality. The environments may support ASP,
PHP, JAVA, MySQL, MSSQL, Adobe Flash,
Cinema 4d studio, and Unity.
b) Develop OVL Platform:
After determining the environment, the students
implement the OVL platform by programming
the learning management tools and the display
content tools inside the OVLs. In addition, they
develop the synchronous and asynchronous
communications tools.
c) Develop OVL Contents:
The students also produce the segments of the
OVL contents such as texts, animations, image,
graphics, video, simulations, electronic questions,
and activities according to the DDOVLs criteria.
They should keep the segments production to
harmonize with the OVL platform.
d) Publish OVL via Internet:
The students specify name of OVLs website
according to the instructional institutions and the
course name. They also provide the necessary
requirements about software and hardware on the
server, such as capacity and security. Finally, they
upload and publish the OVLs via the internet.
363
7.5 Experiment and Implementation Phase:
7.6.1 Conduct Formative Evaluation
a) Conduct OVL Technical Tests
Formative Evaluation for Every Part of

Contents
The OVLs after publishing become an initial

instructional product. Therefore, the students
should conduct technical testing such as interface
and accessibility, navigation, robustness,
database connection, and performance. Based on
these results, they should fix the technical errors
of the initial OVLs.
b) Make Preliminary OVL Experiments with
some Learners
The students perform a preliminary experiment of
the OVLs to a few learners in order to receive
early feedbacks in regard to easy handling,
difficulties, notes and mistakes. In light of the
results, we can make necessary reforms and
become a final version of the OVLs.
c) Make Training of OVL
Training on how to use such new technological
resources contributes to increase the OVLs
efficiency. Thus, they conduct training for the
teachers about how to use the tools and functions
and online teaching skills, and preform training
for the learners about how to access, navigate, use
the tools and functions, and to conduct their
assignments electronically.
d) Apply Actual Learners
In this step, the teachers can use the OVLs
actually in teaching in their classrooms as an
instructional technology tool to start an online
teaching session with the learners, to make
discussion, to present replies for the learners
inquiries, to receive assignments from the learners,
and to offer feedbacks from the performance of
the learners.
7.6 Evaluation Phase:
Evaluation phase presents the students main
indicators about learning inside the OVLs and
quality of the designing process as a whole of the
OVLs. They need to make improvements
permanently in systematic process based on the
results of the evaluation. At this phase, they
would use two types of evaluation.
The teachers conduct a formative evaluation

during the actual learning process inside the
OVLs.Therefore, they should implement it as
electronic questions and questionnaires in every
segment of the OVL contents. The results of this
evaluation provide the following feedbacks with
regard to: revisions and modification of the OVL
contents to the students, the learners progress in
achieving the objectives inside the OVLs to the
teachers, and the level of performance to the
learners in order to motivate them in continuing
in their learning process.
Formative Evaluation for Every Phase
The students also perform a formative evaluation
after each phase during the design process under
supervision of the educational technology
professors to improve the DDOVLs model. They
can use some tools such as evaluation forms,
reports, and checklists. Figure 5 shows an
example of the evaluation form for design
phase.
7.6.2 Conduct Summative Evaluation:
Summative Evaluation for OVL Learning
Process
The students conduct a summative evaluation at
the end of learning, which provides one of the
main indicators about the learning process, the
learners level, and teachers performance inside
the OVLs. They can use some evaluation tools
such as electronic tests and questionnaires.
Summative Evaluation for OVL Designing
Process
The students need to detect the negative and
positive aspects of the whole of the DDOVL
model at the conclusion of the project. They use
some evaluation tools like evaluation forms,
questionnaires and interviews with the teachers
and the learners. They carry on such evaluation
under supervision of the educational technology
professors. The results would improve the OVL
design process continuously.
364
Figure 5. Formative evaluation Form for design phase
7.7 Features of DDOVLs Model

In the proposed DDOVLs model, we skip the
analysis phase since the students in the
faculty of education do not need to make analyses
before DDOVLs. First of all, analysis is a difficult
phase since they need some experience in actual
classroom. In addition, the target learners
characteristics are actually defined by the usual
courses in the schools and universities. Our model
also provides the teachers and the learners with
enough technological training for using the OVLs.
Thus, they do not need the learner and teacher
analysis. The instructional designer and the actual
teacher can specify the content parts like skill,
idea, and experiment directly according to their
experience in teaching the courses on the OVLs.
It means they do not need course analysis, too. On
the other hand, we add the pre-design phase as the
first one. This would make good inputs for the
designing phase. Our model also focuses on the
criteria phase because the criteria are very
important to guide them for DDOVLs with high
efficiency. The professors of the educational
technology can prepare or select any criteria in
this phase.
We divide the design phase into two parts. The

OVLs platform is a fixed part in any OVLs, which
contains a group of tools to make management of
learning and to provide the learners with multi
communications with the teachers and the
contents. On the other hand, the OVLs contents
are variable parts according to the courses. The
OVLs contents are divided into the segments,
each of which belongs to one experiment, skill, or
idea to improve learning. Such segments also
have possibility for reuse in another OVLs. These
parts would make the model general and suitable
to diverse fields.
In the development phase, we do not specify
application development environments to
produce the OVLs so that they can select anything
freely. In experiment and implementation phase,
we claim two ways of experiment as the technical
tests and the preliminary test. The results of the
evaluation phase would make the students
improve the OVLs in systematic process. Our
model components in all phases offer the students
the sequential steps to make them simple and
effective.
365
8. SUGGESTION CRITERIA FOR DDOVLs

We also present a list of criteria for DDOVLs to
be easier in teaching our model directly to the
students. The criteria aim to reach modern
educational and technical principles by providing
guidelines. Consequently, we make a derivation
of some elements of the criteria from the previous
studies regarding to the VLs [38], [39], [40], [41],
[42], [43], [44], and criteria for e-learning
technologies, qualification, and evaluation [45],
[46], [47], [48], [49], [50], [51], [52], [53]. In
addition, other elements of the criteria are picked
up from designing online electronic courses,
developing e-contents, evaluating instructional
websites [54], [55], [56], [57], [58], [59], and [60],
and general aspects of educational technology
[61], [62], [63], [64], [65], and [66].
Although there are diverse types of the previous
studies, we divide the elements of the criteria into
two parts as described in Table 1 so that they can
understand the criteria easily. The one part is
mainly related to educational rules and theories.
Another part is technical rules related to the
computer software. Our criteria contain diverse
specific elements related to the OVLs. Thus, such
elements are underlined in the following sections.
Table 1. Elements of criteria for DDOVLs
DDOVLs Criteria
Educational Criteria
Technological Criteria
a)
b)
c)
d)
e)
a)
b)
c)
d)
e)
f)
g)
h)
i)
j)
Objectives
Contents
activities
Evaluation
Feedback
Online virtual lab platform

Text
Image and graphics
Sound
Video and animation
Navigation
Simulation
Interactivity
Learning administration
Publishing and technical support
8.1. Educational Criteria for DDOVLs

a) Objectives
They should be specified clearly.
They should be related to the contents.
They should be formulated in the behavioral
form.
They should be measurable.
They should be appropriate for the target
learners.
They should be shown at the beginning of each
content.
They should be described as not learning activity

but learning outcomes.
They should be arranged in a sequence according
to expectations of the learning outcomes.
b) Contents
They should be directly related to the educational
objectives.
They should be sufficient.
They should be up to date.
They should be objective.
They should be easily understood, clearness and
succinctness.
They should be accurate in scientific and
linguistic.
They should be organized logically and
sequentially.
They should be designed according to selflearning strategy.
They should be divided into small and gradual
learning steps.
They should be involved in summaries which
describe the main ideas.
They should be classified according to the types
of the contents like practical or theoretical ones.
They should be reviewed by subject-matter
experts such as the teachers or the staffs.
They should be rich with appropriate multimedia.
They should be displayed in various forms, such
as animations, simulations, gures, pictures, and
videos.
They should be supported linear and nonlinear
presentation styles.
They should be clearly specified by additional
information linked to other websites.
Their copyrights should be specified.
c) Activities
They should be related to the OVL objectives.

They should be suitable for the OVL contents.
They should be defined clearly.
They should be designed in a learner-centered
way.
They should send feedbacks to the learners.
They should provide the learners with the time
plan about online and offline learning such as
starting time, duration, and deadline.
They should be included in each contents part.
They should support cooperation among the
learners in a synchronous and asynchronous way.
366
d) Evaluation
It should be related to the educational objectives
in the OVL contents.
Its styles should be appropriate to the OVL
objectives.
It should be covered all the OVL contents.
It should be clearly stated and understandable.
It should have a variety in question types.
Its questions should be ordered by difficulty
level.
Its questions should contain texts, image,
graphics, sound, videos or animations related to
the simulation.
It should provide continuous formative
evaluation during learning.
It should provide summative evaluation after
learning.
e) Feedback
It should be provided immediately.

It should be not general but specific.
It should be suitable for the learners level.
It should be linked to the objective of the
questions.
It should be provided in several shapes such as
texts, images, graphics, sound, video, and
animation.
It should guide the learners to the correct answer
in the case of failure.
It should support the learners with additional
information to the correct answer.
It should give the learners prioritized
information about how their performances do or
do not meet the objectives so that they can
understand how to improve their future
performance.
It should be balanced in terms of the amount and
timing to make it most effective.
It should provide sufficient contrast between the

text or images and the background.
It should be divided into appropriate categories
It should provide multiple channels of
communication
It should provide learning management tools.
It should provide a display function for the
contents.
It should be easy to update the contents.
b) Text
Text-based content should be limited.
It should be clear and readable.
Its font should be selected in appropriate styles
and sizes.
Its headlines should be structured by the font size
hierarchically.
Its titles should be simple and expressive.
It should be formulated in short sentences and
simple composition.
It should be formulated in not passive but active
forms and not contain interrogative forms.
Its color should be suitable compared to the
background.
c) Image and Graphics
They must be original and relevant to the
contents.
They should be clear and not be crowded.
Their color should be appropriate to reality.
Their contrast and brightness should be balanced.
Their digital effects should be related to real and
not use visual tricks to hide from the truth.
They should be highlighted to emphasize the
meaning.
They should show real dimensions.
Their design should be suitable to ask the
questions.
Their quality should be facilitated perception.
8.2. Technological Criteria for DDOVLs
d) Sound
a) OVL Platform
Its tools should be easy and clear to use and

understand.
It should be consistent and provide the learners
with stable operation.
Its design should be artistic and creative.
It should provide search functions.
Its hyperlinks should use the words that clearly
identify where to lead.
The number of it should be limited.

It should be used to make clear the meaning.
It should serve a clear purpose.
It should be pure and not contain echo and noise.
It should be a synchronize audio commentary
with the display.
It should be properly pronunciations.
Its digital effects should be realistic and related
to the contents.
367
It should be high quality as much as possible.

e) Video and Animation
They should serve a clear purpose.
They should use at the necessary time.
Their design should be suitable to ask the
questions.
They should synchronize with audio
commentary if needed.
They should be related to the achievement of the
objective of the OVL contents.
They should display with some control tools
such as stop, repeat, sound, pause, and screen
size.
They should be high quality as much as possible.
They should work to upgrade the level of critical
thinking.
Their speed should be designed according to
specific tasks.
They should increase learning desire.
f) Navigation
It should be easy to move.
Its labels should be easy to understand and
express what they contain.
Broken links should be avoided.
Exit links should be provided at any time.
Back links to the OVL platform should be
provided from every page.
Its links should use graphic buttons.
Main navigation should be stable.
Its links should be consistent and not crowded.
Its structure should be organized logically by the
appropriate labels such as categories, main links,
and sub-links.
Its links should be classified by category, color,
providing links in a new page or using in the
same page.
It should avoid orphan pages.
It should provide the learners with enough
orientation by diagram, mind map, flow chart,
instructional image and graphics.
g) Simulation
It should be related to the contents objectives.
It should provide a virtual environment similar
to the practical real objects.
It should represent the theoretical idea easier to
understand.
Its images, graphics and theoretical idea should
show the meaning as far as possible.
It should be creative.
It should provide interactive environment.
It should harmonize with computer software
used in the producing.
h) Interactivity
Its multiple forms should be provided such as the
teacher to the learners, the learner to the learners,
and the learner to the contents.
It should be provided in synchronous and
asynchronous ways.
It should provide the learners with full control.
i) Learning Administration
It should provide a direct login to the OVLs.
It should provide information about the OVLs
subject and courses.
It should provide announcements such as next
skills, encouragement for participation, time of
contact, guides, results, and award, etc.
It should provide information about new and
emerging technologies such as Wikis, blogs,
collaborative content development software, and
video conferencing software, etc.
It should offer functions to upload and download
files.
It should be balanced between cost and learning
outcomes.
It should provide tools for assessing performance
in the OVLs.
It should provide tracking function for learning.
j) Publishing and Technical Support
The OVLs requirements about software and
hardware should be specified.
The server of the OVLs should have enough
space in consideration of future expansion.
The OVLs domain via the internet should be
simple and suitable content types.
The OVLs pages should be downloaded quickly.
The OVLs should be controlled remotely from
the host institution.
The testing of the OVLs should be validated.
The OVLs should be robust and sustainable to
handle inadvertent damage by the students or
malicious programs.
The OVLs should be easy to maintain in terms of
routine tasks like back-up.
The OVLs should provide accessibility in
anytime and anywhere.
368
The OVLs should supply technical supports for

the learners.
The OVLs should have information about
copyrights of educational institutions.
8.3. Features of DDOVLs Criteria
We have suggested the DDOVLs criteria to
provide modern guidelines for the students in the
faculty of education for designing and developing
process of the OVLs. Our criteria contain some
new areas of elements related to the OVLs and
virtual learning environments. First of all, the
elements related to simulation represent the vital
components of the OVLs with regard to
experiments, skills, relationships, laws, and idea
during the designing process. Secondly, the OVL
platform criteria provide the framework about
communication opportunities, and display of the
OVL contents. In addition, the criteria related to
the OVL learning administration are standard
elements to provide the students with good
management functions for learning inside the
OVLs.
On the other hand, we add a new elements
suitable to the OVLs to grow the main elements
referred to educational criteria such as objectives,
content, activities, evaluation, and feedback.
These elements would make good guideline for
designing the OVL contents. Furthermore, we
pick up the criteria elements which belong to
diverse types of multimedia such as texts, image,
graphics, sound, video, and animation to become
suitable to DDOVLs. We also suggest the criteria
for publishing the OVLs to guide the students
without technical errors.
9. CONCLUSION
In this paper, we proposed the instructional design
model for designing and developing the OVLs in
order for the students in the departments of
educational technology and in the faculties of
education to improve their skills with stable form
and components in the near future schools and
universities. We made the analyses of the
previous studies about the OVLs in many fields
and the previous instructional design models.
Based on the results of the analyses, we proposed
the instructional design model of DDOVLs
focused on e-learning and virtual learning
components suitable in modern educational
systems. The proposed model contains six phases,
pre-design, criteria, design, development,

experiment and implementation, and evaluation
which lead to produce the OVLs step by step.
In addition, we suggested the list of criteria for
DDOVLs to provide the guidelines for the
students. We had made the literature reviews
related to virtual labs, e-learning, online
technological resources, online courses and
instructional websites besides miscellaneous
aspects of educational technology. Depending on
the reviews, we made a derivation of the
DDOVLs criteria and organized them under two
parts: educational criteria and technological
criteria, which consist of 15 categories and 138
elements. These criteria would be used by the
professors of educational technology in the
criteria phase when teaching DDOVLs model to
the students.
In the near future, we will apply the proposed
model to the actual students in the both types of
the students and valid the effectiveness in diverse
fields in the actual educational institute. In
addition, we will develop an OVL template to
reach a pilot system for the OVLs and experiment
it through practical use.
ACKNOWLEDGEMENT
We greatly appreciate the support of the Egyptian
Government Scholarship in ministry of higher
education and scientific research. We thank them
for their priceless contribution of knowledge to
this research.
10. REFERENCES
[1]. K. C. Chu.: what are the benefits of a virtual laboratory
for student learning?. HERDSA Annual International
Conference12-15 July, Melbourne (1999).
[2]. Tatli, Z., Ayas, A.: virtual chemistry laboratory: effect
of constructivist learning environment. Turkish Online
Journal of Distance Education, vol.13 (2012).
[3]. Herga1, N. R., Dinevski, D.: Virtual Laboratory in
Chemistry Experimental Study of Understanding.
Reproduction and Application of Acquired Knowledge
of Subjects Chemical Content. Organizacija, vol. 45, pp.
108-- 116 (2012).
[4]. Lampi, E.: The Effectiveness of Using Virtual
Laboratories to Teach Computer Networking Skills in
Zambia. PhD thesis, Virginia Polytechnic Institute and
State University (2013).
[5]. Bean, L. J ., Fridovich-Keil, J., Hegde ,M., Rudd ,M.
K. , Garber, K.B.: The virtual diagnostic laboratory: a
new way of teaching undergraduate medical students
about genetic testing . Genetics in Medicine 13, 973-977(2011).
369
[6]. Abu Jalal, R. A.: The effect of using the virtual
laboratory on the teaching of mathematics at the basic
education stage, Unpublished Master Thesis, Faculty of
Specific Education, Mansoura University (2009).
[7]. Babateen, H. M.: The role of Virtual Laboratories in
Science Education. In: International Conference on
Distance Learning and Education IPCSIT, vol. 12, pp.
100--104, IACSIT Press, Singapore (2011).
[8]. Harms, U.: Virtual and remote labs in Physics education.
Second European Conference on Physics Teaching in
Engineering Education, Budapest (2000).
[9]. Carnevali, G., Buttazzo, G.: A virtual laboratory
Environment for real-time Experiments. In :The IFAC
Symposium on Intelligent Components and Instruments
for Control Applications, SICICA , pp. 39--44 ,a vireo,
Portugal (2003).
[10]. Prieto-Blzquez, J., Garcia-Tora, I.,
HerreraJoancomart, J., Guerrero-Roldn, A.-E.: Virtual
Laboratory Ontology for Engineering Education. In:
Proc. 2008 of the 38th Annual Frontiers in Education
(FIE).pp. S2F-1--S2F-6, Saratoga Springs New York
(2008).
[11]. Sun, O., Lin, Y., Yu, C.: A study on learning effect
among different learning styles in a Web-based lab of
science for elementary school students. Computers &
Education, vol.50 (4), pp. 1411--1422 (2008).
[12]. Tatli, Z., Ayasb, A.: Virtual laboratory applications in
chemistry education. Social and Behavioral Sciences,
vol. 9, pp. 938--942(2010).
[13]. Muhamada, M., Zamanb, H. B., Ahmad, A.: Virtual
Biology Laboratory (VLab-Bio): Scenario-Based
Learning Approach, Social and Behavioral Sciences. vol.
69, pp. 162--168 (2012).
[14]. Bajpai, M.: Developing Concepts in Physics through
Virtual Lab Experiment: An Effectiveness Study, An
International Journal of Educational Technology,
Techno LEARN, vol.3 (1), pp. 43--50(2013).
[15]. Metrailler, Y. A., Reijnen, E., Kneser, C., Opwis, K.:
Scientific problem solving in a virtual laboratory:
comparison between individuals and pairs. Swiss
journal of psychology, vol. 67, pp. 71--83(2008).
[16]. Tarouco, L., Gorziza, B., Correa, Y., Amaral, E.M.H.,
Muller, T.: Virtual Laboratory for Teaching Calculus:
an Immersive Experience. In: Proc. IEEE Global
Engineering Education Conference (EDUCON), vol. 1,
pp. 774 -- 781. Berlin, jarmany (2013).
[17]. Kantzavelou, I.: a Virtual Lab Model for an Introductory
Computer Science Course. FACTA UNIVERSITATIS
(NIS) SER. ELEC.ENERG. vol. 18(2), pp.263-274(2005).
[18]. Jara, C. A., Candelas, F. A., Puente, S. T., Torres, F.:
Hands-on experiences of undergraduate students in
Automatics and Robotics using a virtual and remote
laboratory. Computers & Education, Vol. 57(4), pp.
2451--2461(2011).
[19]. Tran, N. H., Tran, H. D., Chiem, P. T., Luu, D. T., Cao,
T. D., Kaskenpalo, P.: CenLavi: Virtual Computer
Network Laboratory. In: Fifth International Conference
Ubiquitous and Future Networks (ICUFN), DA NANG,
pp. 523 -- 528 , da nang (2013).
[20]. Pati, B., Misra, S., Mohanty, A.: A Model for Evaluating
the Effectiveness of Software Engineering Virtual Labs.
In: IEEE International Conference Technology

Enhanced Education (ICTEE), IEEE Xplore, pp.1--5
Kerala (2012).
[21]. Yonghua, Z., ShunHong, P., Zhiling, D., Hong, Y.: An
Interactive and Expansible Virtual Laboratory
Environment for Hardware Chips Application
Experiment. Education Technology and Computer
Science, vol. 3, pp. 302 -- 306(2009).
[22]. Rabulets, A.: Systems Engineering Principles of Virtual
Linguistic Laboratories. MONDILEX Second Open
Workshop Kyiv, Ukraine, 24 February, 2009 , pp. 18 -23(2009).
[23]. Zhao, Y., Liu, X., Tian, F., Tian, F.: The Implementation
of Virtual Studio Laboratory Based on Web. In: Proc.
2012 2nd International Conference on Computer and
Information Application (ICCIA 2012), pp. 1167--1170,
Atlantis Press, Paris, France (2012).
[24]. Surez, J., Sancho, F.: A Virtual Laboratory for the
Study of History and Cultural Dynamics. Journal of
Artificial Societies and Social Simulation, vol.14 (4),
pp.1--26 (2011).
[25]. Garn, A., Tusell, F., Unzueta, A.: Enhancing Statistics
Teaching with a Virtual Lab A Case Study of Seamless
Local and Remote Computing. Journal of Information
Technology and Application in Education, vol. 2 (3),
pp.117--124 (2013).
[26]. http://modelscience.com/products.html?ref=home&link
=chemlab
[27]. http://www.crocodile-clips.com/en/Crocodile_Physics/
[28]. http://www.softpedia.com/get/Others/HomeEducation/Virtual-Chemistry-Lab.shtml
[29]. Ahmed, M. E., & Hasegawa, S.: An Instructional Design
Model for Designing and Producing Online Virtual Labs
for Educational Technology Students. In: Proc. the
Third International Conference on E-Learning and ETechnologies in Education (ICEEE2014), pp. 191-205.
Malaysia (2014).
[30]. Rajendran, L., Veilumuthu, R., Divya. J.: A study on the
effectiveness of virtual lab in E-learning. In: (IJCSE)
International Journal on Computer Science and
Engineering, Vol. 2 (6), pp. 2173--2175(2010).
[31]. Chan, C., Fok, W.: Evaluating learning experiences in
virtual laboratory training through student perceptions:
a case study in Electrical and Electronic Engineering at
the University of Hong Kong. Engineering Education,
vol. 4(2), pp.70--75(2009).
[32]. Smaldino, S., Russell, J., Heinich, R., & Molenda, M.:
Instructional technology and media for learning. Upper
Saddle River, New Jersey: Pearson Merrill Prentice Hall
(2005).
[33]. http://en.wikipedia.org/wiki/ADDIE_ModelMorrison
[34]. Morrison, G.R., Ross, S.M., Kalman, H.K., & Kemp,
J.E.: Designing effective instruction (6th ed.). Hoboken,
NJ: John Wiley & Sons, Inc. (2011).
[35]. .Dick, W., Carey, L., Carey, J.O.: The systematic design
of instruction (7th Ed.).Upper Saddle River, New
Jersey: Pearson Merrill (2009).
[36]. Alessi, S. M., Trollip, S. R.: Multimedia for Learning
(3rd Edition), Pearson Allyn & Bacon (2001).
[37]. http://www.umich.edu/~ed626/Gerlach_Ely/ge_main.h
tm
[38]. Cagiltay, N. E., Aydin, E., Aydin, C.C., Kara, A.,
Alexandru, M.: Seven Principles of Instructional
370
Content Design for a Remote Laboratory: A Case Study
on ERRL. IEEE Transactions on Education, vol. 54(2),
pp. 320-327 (2011).
[39]. Padman, V., Memon, N.: Design of a Virtual Laboratory
for Information Assurance Education and Research. In:
Proc. 2002 IEEE Workshop on Information Assurance
and Security, pp.1--7 United States (2002).
[40]. Achuthan, K. , Sreelatha, K.S. , Surendran, S. ,
Diwakar, S. ,
Nedungadi, P. ,
Humphreys, S. ,
Sreekala, S.C.O. , Pillai, Z. , Raman, R. , Deepthi,
A. , Gangadharan, R. , Appukuttan, S. , Ranganatha,
J. , Sambhudevan, S. , Mahesh, S.: The VALUE @
Amrita Virtual Labs Project Using Web Technology to
Provide Virtual Laboratory Access to Students . In:
2011 IEEE Global Humanitarian Technology
Conference, pp. 117--121 Seattle, WA (2011).
[41]. Hristov, G., Zahariev, P., Bencheva, N., Ivanov, I.:
Designing the Next Generation of Virtual Learning
Environments Virtual Laboratory with Remote Access
to Real Telecommunication Devices. In: Proc. EAEEIE
Annual Conference, pp.139--144 Chania (2013).
[42]. Li, L., Zheng, Y., Zhong, S.: Research on Web-based
Virtual Laboratories. In: 2010 International Conference
on Computer, Mechatronics, Control and Electronic
Engineering (CMCE), Vol.1, pp. 478--481 Changchun
(2010).
[43]. Melkonyan, A., Gampe, A., Pontual, M., Huang, G.,
Akopian, D.: Facilitating Remote Laboratory
Deployments Using a Relay Gateway Server
Architecture. IEEE Transactions on Industrial
Electronics, vol. 61(1), pp. 477--485 (2014).
[44]. Besada-Portas, E., Lopez-Orozco, J.A. , de la Torre, L. ,
de la Cruz, J.M.: Remote Control Laboratory Using EJS
Applets and Twin CAT Programmable Logic
Controllers . IEEE Transactions on Education, vol.
56(2), pp. 156 - 164 (2013).
[45]. Shee, D. Y., Wang, Y.: Multi-criteria evaluation of the
web-based e-learning system: A methodology based on
learner satisfaction and its applications. Computers &
Education, vol .50(3), pp. 894--905(2008).
[46]. Mchichi, T., Afdel, K.: Exploiting Web 2.0 technologies
in promoting learning activities: e-learning Web 2
platform .In: International Conference on Education and
e-Learning Innovations, pp. 1--5 Sousse (2012).
[47]. Hus, C., yeh, y., yen, j.: Development of design criteria
and evaluation scale for web-based learning platforms.
International Journal of Industrial Ergonomics, vol.39
(1), pp. 90--95 (2009).
[48]. Ardito, C., De Marsico, M., Lanzilotti. R., Levialdi, S.,
Roselli, T., Rossano, V., Tersigni, M.: Usability of ELearning Tools. In: Proc. working conference on
advanced visual interfaces, pp. 80--84 (2004).
[49]. Holt, D., Segrave, S.: A creating and Sustaining Quality
E-learning Environments of Enduring Value of
Teachers and learners. In G.Crisp, D.Thiele, I.Scholten,
S.Barker and J.Baron (Eds), Interact, Integrate, Impact.
In: Proc. 20th Annual Conference of the Australasian
Society for Computers in Learning in Tertiary
Education. ASCILITE, pp. 226--235, Adelaide,
Australia (2003).
[50]. Jung, I.: The dimensions of e-learning quality: from the
learners perspective. Educational Technology Research
and Development, vol.59 (4), pp.445--464 Association
for Educational Communications and Technology

(2010).
[51]. Wu, I., Chen, W.: Evaluating the E-Learning Platform
from the Perspective of Knowledge Management: The
AHP Approach. Journal of Library and Information
Studies, vol. 11(1), pp 1--24 (2013).
[52]. Pike, J., Huddlestone, J.: E-learning Instructional
Design Guidelines. Human Factors Integration Defence
Technology Centre (2006).
[53]. Thomas, R.: Interactivity & Simulations in e-Learning.
MultiVerse
Solutions
Ltd
(2001),
http://www.jelsim.org/resources/whitepaper.pdf
[54]. Bykzkan, G., Ertek, G., Arsenyan, J.: Evaluation of
e-learning web sites using fuzzy axiomatic design based
approach. International Journal of Computational
Intelligence Systems, vol .3 (1), pp. 28--42 (2010).
[55]. Food and Agriculture Organization of the United
Nations:
http://www.fao.org/docrep/015/i2516e/i2516e.pdf
[56]. elik, S.: Development of Usability Criteria for ELearning Content Development Software, Turkish
Online Journal of Distance Education-TOJDE, vol.13
(2) (2012).
[57]. Teng, X., Muramatsu, B., Zhang, J. W., Teng, X., Tront,
J. G., McMartin, F.: Implementation of Quality
Evaluation for Web-based Courses and Digital Learning
Resource .In: Proc. 3rd International Conference on
Web-based Learning, pp. 1--8,Beijing, China (2004).
[58]. http://elearning.typepad.com/thelearnedman/ID/evaluat
ingcourses.pdf
[59]. Hasan, L.: Evaluating the Usability of Educational
Websites Based on Students' Preferences of Design
Characteristics. International Arab Journal of eTechnology, vol. 3(3), pp.179--193 (2014).
[60]. Eristi, S. D., Sahin-Izmirli, O., Izmirli, S., Firat, M.,
Haseski, H. I.: An Evaluation of Instructional Website
Designs from the Perspective of Visual Perception
Theories: A Study on Scale Development.
Contemporary Educational Technology, vol.1 (4), pp.
348--366(2010).
[61]. Southern Regional Education Board (SREB):
Evaluation Criteria for SREB-SCORE Assets. (2008)
http://publications.sreb.org/2008/08T01_SREB_SCOR
E_Assets.pdf
[62]. Kinicki, A., kreitner, R.: organizational Behavior: key
concepts, skills &best practices. McGraw-Hill Irwin
(2006).
[63]. Ambrose, S. A., bridges, M.W., Dipietro, M., Lovett,
M.C., Norman, M.K.: How Learning (2010).
[64]. Academy of Medical Royal Colleges: standards and
criteria for CPD Activities: A Framework for
Accreditation(2012)
http://www.aomrc.org.uk/publications/statements/doc_
details/9448-standards-and-criteria-for-cpd-activities-aframework-for-accreditation.html
[65]. Southern Regional Education Board (SREB):
Guidelines for Professional Development of Online
Teachers.(2009)
http://publications.sreb.org/2009/09T01_Guide_profde
v_online_teach.pdf
[66]. Calverley, G., Childs. M., Schnieders, L.: Video for
Education. The Association for Learning Technology,
Vol.1 (Distribution Edition), Oxford (208).
371
Applications of the Rotating Calipers to Geometric Problems in Two and Three

Dimensions
Godfried T. Toussaint
New York University Abu Dhabi
United Arab Emirates
gt42@nyu.edu
ABSTRACT
2 THE ROTATING CALIPERS IN 2-D
A paper published in 1983 established that the rotating
calipers paradigm provides an elegant, simple, and yet
powerful computational tool for solving several twodimensional geometric problems. Since then the
rotating calipers have been extended to three
dimensions, and have been applied to many new
problems. In the present paper the history of this tool is
reviewed, and stock is taken of the rich variety of
computational problems and applications that have
been tackled with it during the past thirty years.
KEYWORDS
Rotating calipers, algorithms, computational geometry,
geometric complexity, computer graphics, computer
vision, robotics, combinatorial optimization, line
fitting, statistics, graphs, thrackles, rotating planes
The elegant and simple algorithm in Shamos

thesis [1], showing that the diameter of a convex
n-sided polygon may be computed in O(n) time in
the worst case, resembles rotating a pair of
calipers through 360, once around the polygon.
This concept is illustrated in Fig. 1, which depicts
a convex polygon, and its two horizontal lines of
support (lower and upper) at vertices pi and pj,
respectively. Note that the lines of support are
directed, as indicated by their arrows. This permits
the specification of the direction of rotation, so
that if the lines are rotated in a clockwise direction
while being pivoted about vertices pi and pj, the
respective angles i and j that the lines make with
vertices pi+1 and pj+1, will decrease.
pj
1 INTRODUCTION
The rotating calipers paradigm constitutes a
powerful, simple, elegant, and computationally
efficient tool that can solve a wide variety of
geometric problems in practice. The basic idea
first appeared in the 1978 Ph.D. thesis of Michael
Shamos, where it was applied to the computation
of the maximum distance between the elements of
a convex set: the diameter of a convex polygon in
the plane [1]. Later I coined the name Rotating
Calipers for this procedure, and generalized it in
a number of ways to solve several other twodimensional geometric problems. In 1983 I
presented some of these results at an IEEE
conference in Athens, Greece [2]. Since then the
rotating calipers paradigm has been generalized
further to solve other problems in two as well as
three dimensions. In the present paper the thirtyyear history of this tool is reviewed.
pj-1
pj+1
pi+1
i
pi
pi-1
Fig. 1. The rotating calipers.
2.1 The Diameter of a Convex Polygon

The diameter of a polygon P is the maximum
distance between a pair of points in P. Since P
contains an infinite number of points, searching all
of them is out of the question. To obtain an
efficient algorithm we need a characterization of
372
the diameter in terms of a finite subset of the
points in P. A simple contradiction argument
shows that the diameter of P is determined by a
pair of vertices of P. If two points of P determine
the diameter and are not vertices of P then the line
segment joining these two points may be increased
in length by moving either point so that it
coincides with a vertex of P. Therefore the
diameter may be calculated by examining only the
distances between all the pairs of vertices of P,
and selecting the maximum distance encountered.
However, this nave (brute force) search algorithm
requires a number of operations that grows as the
square of the number of vertices in the polygon. A
more fruitful characterization narrows the set of
candidates to be searched down to linear size.
Several approaches have been tried in the past in
order to speed up diameter-finding algorithms [3],
and some characterizations have proved to be
incorrect [4]. Several hill-climbing algorithms
published are not guaranteed to yield the diameter
because they assume that the distance between a
vertex of a convex polygon and the remaining
vertices scanned in order, is a unimodal function,
when in fact such a distance function may have
(n) local maxima [4]. Indeed, a convex polygon
may have (n2) pairs of vertices that are local
maxima of their distance functions. However, a
valid characterization of the diameter was
obtained by Shamos [1] via the pairs of vertices,
such as pi and pj in Fig. 1. These vertices are
antipodal, meaning that they admit parallel lines
of support. Shamos showed that the diameter of a
polygon is determined by two of its antipodal
vertices. Furthermore, a polygon with n vertices
has O(n) antipodal pairs, assuming that all the
vertices of P that have an angle of 180 have been
removed, which is a straightforward matter. The
rotating calipers provide a simple O(n) time
procedure for searching all the antipodal pairs to
find the maximum. The idea is to first place a pair
of parallel lines of support in any orientation, say
horizontal, as in Fig. 1, and then rotate the lines,
while keeping them as support lines of the
polygon, until they are horizontal again. Such a
procedure will visit all pairs of antipodal vertices.
The crucial observation that makes this seeming
infinite continuous process discrete and finite is
that the rotation process can hop from vertex to
vertex. Observe in Fig. 1, that as the two lines of

support rotate, the vertices pi and pj maintain their
antipodality property until one of the lines lies
flush with an edge of P. In Fig. 1, j is smaller
than i, and thus the line advances from pj to pj+1
identifying the next antipodal pair pi and pj+1. At
each step all that is needed is a comparison of two
angles to determine which of the two is smaller,
which can be done in constant time.
2.2 The Width of a Convex Polygon
The width of a polygon P is the minimum distance
between a pair of parallel lines of support of P. As
with the diameter definition, to obtain an
algorithm, the width must be characterized by a
finite subset of the lines of support that can be
searched efficiently. The following property yields
such a characterization. Let pi and pj be two
vertices of the convex polygon that admit parallel
lines of support, and have internal angles less than
180, as in the example of Fig. 1. If no edge of P
lies on the support lines, there exists a preferred
direction of rotation for the support lines such that
their separation distance decreases. This implies
that the width of the polygon is characterized by a
vertex and an edge (a vertex-edge pair) that are
antipodal, i.e., such that one of the lines of support
lies flush with an edge, as shown in Fig. 2. This
characterization of the width of a convex polygon
found application to the segmentation of plane
curves in the work of Ichida and Kiyono [5] (see
also [6]). The characterization has also led to
several algorithms for computing the width of a
polygon. A useful property in this regard is the
fact that for a line that contains any edge of a
convex polygon P, the perpendicular distance
between the line and the vertices of P, as they are
traversed in order, defines a unimodal function
[7]. Kurozumi and Davis [8], and Imai and Iri [9],
independently proposed algorithms for computing
the width of a convex polygon by visiting each
edge of the polygon, and for each edge searching
for the vertex furthest from it (in a perpendicular
sense). Since this distance function is unimodal
the algorithms in [8] and [9] apply binary search
to locate these vertices, for each edge of P. This
approach results in algorithms with O(n log n)
worst-case time complexities.
373
pj
pj-1
j
pj+1
pi+1
i
pi
pi-1
Fig. 2. The rotating calipers and the width of P.
Houle and Toussaint [10] show that an O(n) time

algorithm is achievable by avoiding binary search
altogether, and using the rotating calipers instead.
To initialize the algorithm, a line of support is
constructed through any edge, such as pi-1 and pi in
Fig. 2, and the vertex furthest from this line is
found (pj in Fig. 2). At each step during the
rotation of the lines, the succeeding edge selected
is the one that makes the smaller angle with its
line of support. This process yields the vertex
opposite the edge in only constant time.
2.3 The Minimum-Area Enclosing Rectangle
The algorithm described in the previous subsection, for computing the width of a polygon with
the rotating calipers, represents an application of
the original calipers that use two parallel lines of
support to compute the diameter, to solve a
different problem. However, the rotating calipers
tool itself has also been generalized in several
ways. One generalization introduced in [1] uses
more than two rotating lines of support. One nice
example uses four lines of support to tackle a
problem that arises in the areas of image
processing and computer vision [11], [12], [13][15] optimal packing and layout problems in
manufacturing [16], and automatic tariffing in
goods traffic [17]. The problem is that of
computing the minimum-area enclosing rectangle
of a convex n-sided polygon. The usefulness of
this rectangle in packing problems is obvious, but
it also has applications to shape analysis. For
example, the rectangularity of a shape may be
measured by the difference in the areas of the
shape and its smallest enclosing rectangle [12].
Freeman and Shapira [16] showed that the
smallest rectangle must have one of its four edges
flush (collinear) with an edge of the polygon, as

illustrated in Fig. 3. The algorithm they propose
involves visiting every edge of P, such as edge [pi1 pi] in Fig. 3, and locating the three associated
extreme vertices that complete the enclosing
rectangle, such as pt, pj and ps in Fig. 3. Their
algorithm inspects all the vertices of P to find
these three extreme vertices, leading to a total
computational complexity of O(n2). If on the other
hand each of these extreme vertices is located in
O(log n) steps using binary search, instead of a
linear scan (which is valid due to the unimodality
property of the distance functions involved [45]),
then the solution may be found in O(n log n) time.
However, the rotating calipers with four lines of
support solves this problem elegantly in O(n) time,
as follows. To initialize the procedure any edge of
the polygon is selected as the base of a candidate
for the smallest enclosing rectangle, and the three
extreme vertices are found by a linear scan of all
the vertices, as in [16]. The rest of the algorithm
proceeds in a manner similar to that used in the
algorithm for computing the width of the polygon,
except that here the four lines of support are
rotated by the smallest of the four angles that the
lines make with their succeeding clockwise edges,
as shown in Fig. 3. In this way every edge of the
polygon generates its corresponding candidate
rectangle as the support lines make a full
revolution around the polygon. Furthermore each
candidate rectangle is generated in constant time,
resulting in a total time complexity of O(n). An
alternate approach to solve this problem in O(n)
time, that uses a data structure known as a star, is
described in [18] (see also [19] for the use of the
star in a different context).
pj
t
pt
ps
s
i
pi
pi-1
Fig. 3. The minimum-area enclosing rectangle.
374
In closing this sub-section it is worth noting that
the rotating calipers have also be used to find
minimum-area enclosing triangles [20], [21],
squares [22], minimum-perimeter enclosures [23],
and the densest double-lattice packing of a convex
polygon [24]. The case of the minimum-area
triangles enclosing a convex polygon is most
similar to the rectangle problem considered in this
subsection. Klee and Laskowski [20] provided the
key characterization of the optimal solution. They
proved that if T is a local minimum among the
triangles that contain the polygon, then at least one
side of T must lie flush with an edge of the
polygon. This property led the authors to propose
an O(n log2 n) time algorithm for finding the
optimal solution. ORourke et al., used the
rotating calipers to solve this problem in linear
time [21]. The rotating calipers used with four
rotating lines can also be used to find the
maximum-area quadrilateral enclosing a convex
polygon. Hosono, Meijer, and Rappaport [25] use
it to compute the visibility graph of a set of nonintersecting translates of the same compact convex
object in the plane.
2.4 The Maximum Distance Between Two
Convex Polygons
The maximum distance between two convex
polygons, P and Q, arises in several applications
including pattern recognition, cluster analysis, and
unsupervised learning [26]. It is defined as the
largest distance determined by one point in P and
another point in Q. As with the diameter problem,
the search for the maximum distance may be
restricted to the vertices of P and Q, denoted by
p1, p2, , pn and q1, q2, , qn, respectively. This
maximum distance between P and Q is given by:
dmax(P,Q) = max{d(pi, qj)}, i, j=1, 2, , n,
(1)
where maximization is done over all i and j, and

d(pi, pj) is the Euclidean distance between pi and
pj. Bhattacharya and Toussaint [27] showed that
two sets of points may be partitioned into eighteen
subsets such that the maximum distance between
the two sets is equal to the largest of the eighteen
diameters of each of these subsets. Furthermore,
the diameter of each subset may be computed with
the rotating calipers algorithm applied to their

convex hulls. If the two sets are convex polygons,
their algorithm runs in O(n) time. However, a
simpler and direct O(n) time algorithm for the case
of convex polygons was later discovered by
Toussaint and McAlear [28]. Their algorithm
follows from the generalization of the notion of an
antipodal pair of points for a single polygon, as
illustrated in Fig. 4.
pi
i
pi-1
pi+1
qj+1
qj-1
j
qj
Fig. 4. The maximum distance between two convex

polygons is determined by an antipodal pair between them.
An antipodal pair between two sets P and Q, is

defined as a pair of vertices pi P and qj Q, such
that they admit parallel lines of support of P and
Q, at pi and qj, respectively, with the added
restrictions that the support lines are oriented in
opposite directions, and each polygon lies to the
right of its support line. Note that both polygons
need not be contained in the parallel strip defined
by the two support lines, as is the case for the
configuration illustrated in Fig. 4. If one polygon
is smaller than the other, or if it lies inside the
other, it may protrude outside of this strip. It is
shown in [28] that the maximum distance between
the two sets is determined by an antipodal pair
between the sets. There are only a linear number
of such pairs, and they can be searched in O(n)
time by rotating the calipers in the same manner as
was done in the diameter algorithm. In Fig. 4, the
pair pi and qj are one candidate pair, and the next
candidate pair is obtained in O(1) time by rotating
the calipers in a clockwise manner by the smaller
of the angles i and j.
375
2.5 Minkowski Sum of Two Convex Polygons
Consider two points r and s in the plane, denoted
by r(xr,yr) and s(xs,ys), specified by their x and y
coordinates. The Minkowski sum (also called the
vector sum) of r and s is the new point t(xt,yt),
where xt = xr+xs and yt = yr+ys. The Minkowski
sum of two convex polygons P and Q, denoted by
P Q, is the set of points obtained by the
Minkowski addition of each and every point in P
with each and every point in Q. The Minkowski
sums of polygons in the plane, and polyhedra in
space, find application in spatial planning
problems in the field of robotics [29], [30]. The
Minkowski sum of two convex polygons, P and Q
may be characterized in terms of their vertices,
thus making it computable. In particular, P Q is
a convex polygon, and has at most 2n vertices,
which are Minkowski sums of the vertices of P
with those of Q. This characterization implies the
following algorithm: first compute all (n2)
pairwise Minkowski additions of the vertices of P
and Q, and then compute the convex hull of the
resulting set. Using an efficient O(n log n) time
convex hull algorithm, such as Grahams
algorithm [31], yields an O(n2 log n) time
algorithm for computing the Minkowski sum.
However, a much faster O(n) time algorithm may
be obtained by exploiting a characterization of the
Minkowski sum in terms of a modification of the
notion of an antipodal pair of vertices. Two
vertices pi P and qj Q are defined as being copodal if, and only if, they admit parallel directed
lines of support of P and Q, at pi P and qj Q,
respectively, such that the support lines are
oriented in the same direction, and each polygon
lies to the right of its support line, as illustrated in
Fig. 5. The following characterization of the
Minkowski sum of P and Q may now be obtained:
the vertices of P Q are the Minkowski sums of
co-podal pairs of vertices of P and Q. This
characterization permits the computation of P
Q by rotating the calipers in the manner as shown
in Fig. 5, where the pair pi and qj is a candidate
pair of vertices to be summed. The subsequent
candidate pair is obtained in O(1) time by rotating
the calipers in a clockwise manner by the smaller
of the angles i and j. Let zk = pi qj denote a
vertex of P Q that has been computed. Then the

succeeding vertex zk+1 = pi qj+1 if j < i, zk+1 =
pi+1 qj if i < j, and zk+1 = pi+1 qj+1 if i = j.
Since there are no more than 2n vertices in P Q
it follows that O(n) time suffices to compute it.
P
pi+1
pi
qj+1
pi-1
Q
qj-1
j
qj
Fig. 5. The Minkowski sum of two convex polygons.
2.6 The Convex Hull of Two Convex Polygons

There exist applications where it is required to
compute the convex hull of two convex polygons.
For example, the divide and conquer approach to
computing the convex hull of a set of n points in
the plane repeatedly (recursively) merges the
convex hulls of two smaller subsets of the points
[32]. A linear time merge step is sufficient to yield
an algorithm for the convex hull of the set that
runs in O(n log n) time. The convex hull of two
convex polygons, P and Q, consists of three types
of edges: edges of P, edges of Q, and edges that
connect vertices of P with those of Q. The latter
edges are called bridges (also common tangents).
In Fig. 6 (a) the dashed line LB that supports
polygons P and Q at vertices pi and qj,
respectively, determines a bridge of the convex
hull of P and Q. The convex hull of two convex
polygons may therefore be computed by rotating
clockwise two directed lines of support (one on
each polygon) oriented in the same direction, such
that each polygon is to the right of its line of
support, as was done for the Minkowski sum
problem. Whenever the two lines of support are
not collinear (overlapping) one of them lies to the
left of the other. For instance, in Fig. 6 (a), line LQ
lies to the left of LP. Later in the process line LP
will lie to the left of LQ. This implies that the lines
376
must, at some time during the rotation, completely
overlap, and whenever they do so they identify a
bridge. Thus, the convex hull of the two polygons
may be obtained by rotating the calipers a full
revolution, and at each step outputting the vertex
of either P or Q, that lies on the leftmost
supporting line. Since each vertex of P and Q is
visited only once, and there are O(n) bridges, O(n)
time suffices for the entire computation.
For example, in Fig. 6 (b) the angles formed by

(pr-1, pr, qs), (pr+1, pr, qs), (pr, qs, qs-1,), (pr, qs,
qs+1,) are all right turns, and therefore pr and qs
determine a bridge.
pr
pr-1
pr+1
qj
LQ LP
LB
qs-1 qs
pi+1
pi+1
pi
qs+1
qj-1
pi
j
pi-1
qj+1
Fig. 6 (b). Finding the bridge of two convex polygons with

one rotating caliper line.
qj
LB
qj-1
Fig. 6 (a). The convex hull of two convex polygons.
An alternate O(n) time algorithm for finding the

common tangents of two convex polygons, that
uses only one rotating line, was described by
Bernard Chazelle in the context of a divide-andconquer algorithm for finding the convex hull of n
points which are first sorted by x-coordinate [33].
Assume the polygon P is left of Q, and refer to
Fig. 6 (b). First the leftmost vertex pi P is found.
Then the vertex qj Q is determined such that the
line defined by pi and qj is tangent to Q, and Q lies
beneath it. This line is the starting position for its
clockwise rotation while remaining a support of Q.
At the next step of this rotation, the line is
advanced either to vertex pi+1 or to vertex qj+1,
depending on which angle of rotation is the
smaller of the two. Throughout the rotation, each
time a vertex of P and another of Q are both on the
rotating line, the vertices are tested in O(1) time to
determine if they constitute a bridge. A bridge has
the property that the four angles determined by pi
and qj and its adjacent vertices are all right turns.
Computing bridges between two convex polygons

have also found application in solving special
cases of the travelling salesperson problem (TSP),
in which there are nested convex obstacles [34].
2.7 Intersecting Two Convex Polygons
Computing the intersection of two convex
polygons is a fundamental operation that occurs in
many applications. For example, the divide and
conquer approach to computing the intersection of
a set of n half-planes, repeatedly merges the
intersections of two smaller subsets of the halfplanes, which are (perhaps unbounded) convex
polygonal sets [35]. A linear-time algorithm for
intersecting two convex polygons, that uses the
slab method, was described by Michael Shamos in
his thesis [1], which leads to an O(n log n) time
algorithm for the half-plane intersection problem.
An alternate O(n) time algorithm was proposed by
ORourke [36]. A transparently clear O(n)
algorithm along with an easy proof of correctness,
that uses the rotating calipers was presented in
[37]. The algorithm exploits the fact that if two
convex polygons intersect, then there exists an
intersection point corresponding to each bridge in
the convex hull of the two polygons, as illustrated
in Fig. 7, where the dashed line LB identifies a
bridge determined by vertices qj Q and pi P,
377
and x denotes the intersection point corresponding
to this bridge. The algorithm in [37] first
determines if the two polygons intersect. If they
do then the rotating calipers are used to find the
convex hull as described in Sub-section 2.6, after
which for each bridge the corresponding
intersection point is found by a simple step-down
procedure along the convex chains searching for
the two intersecting edges.
LB
detected each time that both support lines are

collinear (overlap each other), as in Fig. 8.
LP
pi
LB
qj
qj+1
qj-1
pi-1
pi+1
pi-1
pi
qj-1
pi+1
Fig. 7. A bridge and its corresponding intersection point x of

two intersecting convex polygons.
2.8 The Critical lines of Support of Two

Convex Polygons
Given two disjoint convex polygons P and Q, the
critical support lines are the two lines that separate
P and Q, such that they are both support lines for
P and Q. In Fig. 8 the two critical support lines, LP
and LQ, are the dashed lines. One line contains
vertices pi+1 and qj, and the other contains pi-1 and
qj-1. Intuitively, the two critical support lines may
be obtained by rotating any line that separates the
two polygons, in clockwise and counterclockwise
directions as much as possible, while maintaining
the separability of the two polygons. Critical
support lines find application to a variety of
problems, some of which are described in the
following subsections. For two convex polygons
they can be computed in linear time using the
rotating calipers in a manner similar to that of
computing the maximum distance, by initially
placing the two parallel lines oriented in opposite
directions such that each polygon is to the right of
its support line, as in Fig. 4. During the rotation
phase of the calipers the critical support lines are
qj
Q
qj+1
LQ
Fig. 8. The two critical support lines (dashed) determined by

two disjoint convex polygons.
2.9 The Widest Separating Strip Between Two

Convex Polygons, and Machine Learning
In the context of pattern classification and
machine learning, two polygons P and Q may be
thought of as regions in a feature space that
enclose points of the training data for a two-class
discrimination problem. Any separating line then
is a linear classification rule that can be used to
classify future patterns (points) depending on
whether they lie on one side or the other of this
line. In order for the classifier to make more
confident decisions it is desired to pick the
separating line that is furthest from the two
polygons. Such a line may be chosen as the
centerline of the widest empty strip that separates
the polygons. Such classifiers are referred to as
large-margin classifiers [38], (also wide separation
of sets [39]) and make up the geometric backbone
of support vector machines [40]. For the case of
two planar convex polygons the widest empty
strip is determined by either one vertex from each
polygon, or by a vertex of one polygon and an
edge of the other. Furthermore, this strip may be
detected when the support lines are oriented in
378
between the directions of the two critical support
lines, and may be found in O(n) time with the
rotating calipers.
The widest empty strip problem is closely related
to the problems of fitting lines to data [41], finding
transversals of sets [42], and linear approximation
of objects [43], [44], all of which have been
solved efficiently using the rotating calipers.
Similar problems occur when data is not linearly
separable. In this setting Aronov et al., [45] use
the rotating calipers to find linear separators that
minimize a variety of different measures of error
for the points misclassified by the separator.
2.10 The Grenander Distance Between Two
Convex Polygons
Ulf Grenander [46] proposed that the distance
between two disjoint convex polygons be
measured by comparing the lengths of the
connecting segments of their critical support lines,
to the lengths of the polygonal chains spanned by
these segments. To be more precise let the critical
support lines be supported at pi and pr in P, and qj
and qs in Q, and refer to Fig. 9.
qj
qj+1
pr
Q
qs
pi+1
pi
Fig. 9. The Grenander distance between two disjoint convex

polygons.
The support vertices of the critical support lines

partition the convex polygons into two polygonal
chains: the inner chains visible from the
intersection point of the support lines, and the
complementary outer chains invisible from the
intersection point. The supporting chords of P and
Q are the line segments that connect qj to pi and pr
to qs, shown as bold red line segments in Fig. 9.

The Grenander distance is defined as the sum of
the lengths of the two supporting chords, less the
sum of the lengths of all the edges of the polygons
that belong to both inner chains. Clearly, once the
critical support lines have been computed with the
rotating calipers, an additional O(n) time suffices
for the computation of the lengths of the chords
and chains, and therefore the Grenander distance
may be computed in O(n) time overall.
2.11 Optimal Strip Separation in Medical
Imaging and Solid Modeling
In several contexts such as medical imaging it is
required to construct a solid model by stitching
parallel polygonal slices together. A problem
arises during the interpolation when the solid
object is bounded by a single contour in one slice,
and two contours in the adjoining slice. The
computational geometric problem that results is
the following [47]. Given two linearly separable
polygons P and Q, and a third convex polygon R,
it is required to compute the separating strip
between P and Q, that covers the largest area of R.
Barequet and Wolfers [47] present a linear-time
algorithm for computing this optimum strip using
the rotating calipers. They also consider the case
when the polygon R is not convex, but in this case
the running time of their algorithm is quadratic in
the size of the input.
2.12 Aperture Angle Optimization for Visibility
Problems in Graphics and Computer Vision
In several disciplines such as computer graphics,
computer vision, robotics, operations research,
visual inspection, and accessibility analysis in the
manufacturing industry, the notion of visibility is
fundamental. Often models assume that a camera
can see in all directions, in effect idealizing the
cameras aperture angle to 360. In a more realistic
model the aperture angle is much smaller than
360. Furthermore, if the camera is movable, it is
desirable to compute the maximum and minimum
aperture angles that the camera may need as it
travels in a constrained space. Let P and Q be two
disjoint convex polygons in the plane. For a given
a point x in P, the aperture angle at x with respect
379
to Q is defined as the angle of the cone with apex
at x, that contains Q, and has its two rays that
emanate from x tangent to Q. The critical lines of
support play singular roles in computing the
extreme aperture angles, and may be efficiently
computed with the rotating calipers [48].
2.13 Wedge Placement Optimization Problems
Wedge placement optimization problems arise in
several contexts such as visibility with bounded
aperture angles, and layout design of parts in stock
cutting for manufacturing. A wedge W may be
thought of simply as an unbounded cone with a
fixed angle at its apex. Given an n-vertex
polygonal region, such as R in Fig. 10, we are
interested in computing the entire region where a
camera (the apex of the cone) with aperture angle
, may be positioned so that it is as close as
possible to R. Such a region is bounded by a
concatenation of arcs (called the wedge cloud)
determined by the apex of the cone as it travels
around R maintaining contact with R. Fig. 10
shows three points on the cloud, xi, xj and xk which
are camera locations with a fixed aperture angle .
The original rotating calipers can be generalized
so that the lines of support form any fixed angle to
each other.
xi
xj
xk
Fig. 10. Wedge fitting and the wedge cloud.
Teichmann [49] used the above generalization of

the calipers to compute the wedge cloud of a
convex polygon in O(n) worst-case time. Note that
to maintain contact with the polygon R, a wedge
must both rotate and translate.
2.14 Nonparametric Decision Rules

In the nonparametric discrimination problem we
are given training data that belong to different
classes, and it is desired to classify new incoming
data into their respective classes. Jean-Paul
Rasson and Grandville [50] proposed a geometric
approach to the design of such a decision rule.
Consider the two-dimensional, two-class problem,
in which the classes are linearly separable, and
refer to Fig. 11. In the training phase of the
classifier the convex hulls P and Q of each class
are computed and stored.
X
qj
pr
qj+1
pi+1
pi
qs
Fig. 11. A geometric non-parametric decision rule.
The decision rule for a new incoming pattern X is

as follows. If X lies in P (or Q) it is classified as
belonging to class P (or Q). If X lies outside both
polygons it is classified to the class associated
with the nearest polygon, where nearest is defined
in terms of the area distance between X and each
of the polygons. More precisely, the distance
between X and Q is the absolute value of the
difference between the area of polygon Q and the
area of the convex hull of Q X. Similarly, the
distance between X and P is the absolute value of
the difference between the area of polygon P and
the area of the convex hull of P X. These areas
are colored dark blue and dark green, respectively.
In this example the dark green area is smaller than
the dark blue area, and therefore X would
classified as belonging to class P. The four lines
that connect X to the polygons are critical support
lines between X and the polygons, where X may be
considered as a degenerate single-vertex polygon,
and may be computed in linear time with the
rotating calipers.
380
2.15 Nice Triangulations and Quadrangulations
of Planar Sets of Points
A convex polygonal annulus is the region in
between two properly nested convex polygons,
such as the blue shaded region consisting of the
polygon Q less the interior of the green shaded
polygon P (Fig. 12). This region admits a very
simple triangulation by means of the rotating
calipers [51]. A triangulation of a polygonal
region is a partition of the regions interior into as
many triangles as possible, obtained by inserting
diagonals between pairs of vertices, without
allowing any edges to properly cross each other.
Initially two parallel lines of support, oriented in
the same direction, are constructed through a pair
of extreme vertices such as the vertices of P and Q
with lowest y-coordinates, such that the polygons
lie to the right of the lines (Fig. 12). As the
calipers are rotated clockwise, the pairs of vertices
that come into simultaneous contact with the lines
of support are connected with an edge. The first
few edges connected by this algorithm are shown
in red in Fig. 12. In general a region admits many
triangulations, some of which may have long
edges, acute triangles, or other properties deemed
undesirable for some applications such as mesh
generation. One attractive property of the
triangulation of the annulus obtained with the
rotating calipers algorithm is that the triangulation
tends to be nice, in the sense that the resulting
triangles tend to be nearly regular. Furthermore,
the method can be applied to more general
problems such as obtaining nice triangulations of
sets of points. One way of doing this is to first
compute all the convex layers of the set [52],
which yields a nested collection of annuli, each of
which can be triangulated with the rotating
calipers [53]. Another method involves first
computing a spiral polygonal chain spanning the
points [54], and then triangulating this spiral with
the rotating calipers [53]. This latter triangulation
also has the added nice property that it is
serpentine, i.e., its dual graph is a chain. In a
similar approach the rotating calipers have been
used to count and enumerate pointed pseudotriangulations using the greedy diagonal flipping
algorithm, by rotating the calipers along the
interiors of two pseudo-triangles [55].
qj+3
Q
pi-2
qj+2
pi+1
i
qj+1
pi
j
qj
Fig. 12. Triangulating a polygonal convex annulus.
2.16 The Rotating Caliper Graph

The rotating caliper procedure, besides solving
many problems directly, has also given rise to a
new geometric data structure called the rotating
caliper graph (RCG) that has found several
applications to the design of efficient geometric
algorithms [56]. Fig. 13 shows a convex polygon
and its rotating caliper graph. The edges of the
RCG consist of edges that connect the antipodal
pairs of vertices of the polygon that are visited
during a full rotation of the parallel lines of
support in the diameter algorithm. The RCG may
also be defined for arbitrary point sets, but in this
case the caliper it only connects the convex hull
vertices of the set. The RCG has several
interesting and useful properties. One of these is
of course that it contains a small linear number of
edges. David Eppstein has exploited this property
to obtain efficient algorithms for maintaining the
RNG itself as well as the width and diameter of
points sets under insertion and deletion operations
of individual points [56]. Another interesting
property of the RCG is that it is a thrackle. John
Conway defined a thrackle as a planar graph
embedding (drawing) with the property that every
pair of edges intersects at a single point [57], [58].
In the case of the RCG each pair of edges contains
an intersection point at either a common endpoint
of the edges, or in their interiors. Since the edges
are straight line segments, this type of thrackle is
called a linear thrackle.
381
Fig. 13. Convex polygon and its rotating caliper graph.
3 THE ROTATING CALIPERS IN 3-D

3.1 The Convex Hull of a Set of Points
The natural generalization of the process of
rotating a line constrained to remain tangent to a
convex polygon in two dimensions, to 3dimensional space, is rotating a plane that remains
tangent to a convex polyhedron. To specify the
direction of rotation uniquely the plane should be
tangent to two points in space, which determine
the axis of rotation of the plane. The simplest
example of such an application is in the
computation of the convex hull of n points in 3-D.
A simple algorithm for this purpose first sorts the
points by increasing x-coordinate, and then
constructs the convex hull incrementally by
inserting one point at a time in the sorted order in
a growing convex hull. To initialize the process
the four points with smallest x-coordinates may be
connected to form a tetrahedron, which serves as
the convex hull of the four points. At every
subsequent step in the execution of this algorithm
we have a convex polyhedron Pi-1 (the convex hull
of the first pi-1 points considered), and the i-th
point pi that lies to the right of Pi-1. The process of
inserting pi is illustrated in Fig. 14, where the zaxis (not shown) is orthogonal to the xy-plane, in
the direction of the viewer. First a plane
orthogonal to the xy-plane that contains pi and is
tangent to Pi-1 at point a, is found. The projection
of this plane onto the xy-plane is shown in Fig. 14
as the line through pi and a. Using this line as the
axis of rotation, the plane is rotated until it collides
with another vertex of Pi-1 (vertex b in the Figure),
thus creating a triangle (pi, a, b), shown shaded in
transparent green. This triangle becomes a face of

the new convex hull, and the line through pi and b
becomes the new axis of rotation for the rotating
plane. This process continues until the rotating
plane arrives at the edge (pi, a) again. The new
faces inserted in the new convex hull form a cone
with apex at pi. Finally, the faces of the previous
convex hull that lie inside this cone and are visible
from pi are deleted, to complete the new convex
hull. The addition of each point to the growing
convex hull can be done in O(n) time leading to an
algorithm with O(n2) time overall [59]. If a divideand-conquer approach is used to compute the
convex hull, a similar approach may be used in
which the rotating plane wraps around two convex
polyhedral. Such a procedure, originally proposed
by Preparata and Hong [60], leads to an algorithm
that runs in O(n log n) time.
y-axis
pi
a
b
a'
x-axis
Fig. 14. Computing the convex hull in 3-D incrementally

using a rotating plane.
The above algorithm for inserting a new vertex

into the 3-dimensional convex hull of points in
space can also be used to solve other problems. In
the hidden-surface problem in computer graphics
the point pi represents a point light source, and the
rotating calipers then yields the portion of a
convex polyhedron illuminated by that point. In
the problem of non-parametric classification the
point pi represents a new pattern to be classified.
Here the volume of the region inside the cone
determined by pi and the portion of the convex
polyhedron visible from pi is a measure of the
distance from pi to the polyhedron, thus
generalizing the 2-dimensional approach of
Rasson and Grandville [50] to three dimensions.
382
3.2 Finding Minimal Enclosing Boxes
A natural generalization of the problem of finding
the minimum-area rectangle that encloses a given
convex polygon, is finding the minimum-volume
box that contains a given convex polyhedron.
However, unlike its two-dimensional rectangular
counterpart, the minimum-volume box bounding a
convex polyhedron need not have one of its faces
flush with a face of the polyhedron. Fig. 15
illustrates the minimum-volume box enclosing a
regular tetrahedron (shaded in blue) in which no
faces of the box are flush with faces of the
tetrahedron. Note however, that in this example
every face of the box is flush with an edge of the
tetrahedron. In fact, ORourke [61] showed that a
box of minimum volume enclosing any convex
polyhedron must have at least two adjacent faces
flush with edges of the polyhedron. This
characterization allows ORourke to design an
algorithm that performs a type of threedimensional caliper rotation starting with every
pair of edges of the polyhedron, that runs on O(n3)
time. The caliper in effect consists of two planes
orthogonal to each other that rotate in unison,
reminiscent of the wedge-placement optimization
problems in two dimensions.
to previous methods, and conclude that their

algorithm is faster and more reliable.
In practice one may be satisfied with an even
simpler algorithm that yields an approximation to
the smallest bounding box. Such a method
proposed by ORourke constructs the minimumvolume enclosing box with the constraint that one
of the faces of the convex polyhedron lie flush
with a face of the box. In this version of the
problem it suffices to project the convex
polyhedron onto every plane determined by a face
of the polyhedron, projecting the polyhedron onto
that plane, and solving the minimum area
enclosing rectangle problem with the planar
rotating calipers. The resulting rectangle then
forms the base of the box, the height of which is
determined by the vertex of the polyhedron that is
furthest from the base. This simple approximation
algorithm runs in O(n2) time. Approximation
algorithms that run in almost linear time exist, but
they are more complicated, and use for example,
principal components rather than the rotating
calipers [63], [64]. Additional optimization
methods for finding bounding boxes and a
discussion of their application to brachytherapy
(internal radiotherapy) may be found in [65].
4 CONCLUSION
Fig. 15. The minimum-volume bounding box of a regular

tetrahedron.
Motivated by the fact that ORourkes O(n3) exact

algorithm is complicated and slow in practice,
Chang, Gorissen, and Melchior [62] proposed an
approach formulated as an unconstrained
optimization problem. They compare their method
This paper has focused on extensions of the

rotating calipers, and their applications to a variety
of geometric problems in the plane and three
dimensions. While the problem of computing the
minimum distance between sets, resembles
superficially the maximum distance problem, it is
conspicuously absent here. This is because the
problem appears difficult to crack with the
rotating calipers [66]. The rotating calipers have
also been generalized to work on two-dimensional
manifolds other than planes, such as spheres and
cones [67]. The reader is referred to the web page
and thesis of Hormoz Pirzadeh [68], for details of
some characterization proofs, and animation
applets that help to visualize several of the 2dimensional algorithms that have been described
in this paper.
http://cgm.cs.mcgill.ca/~orm/welcome.html
383
4 ACKNOWLEDGEMENTS
This research was supported by a grant from the
Provost's Office of New York University Abu
Dhabi, through the Faculty of Science, in Abu
Dhabi, The United Arab Emirates.
5 REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
M.I. Shamos, Computational geometry, Ph.D. thesis,

Yale University, 1978.
G. T. Toussaint, Solving geometric problems with the
rotating calipers, Proceedings of IEEE MELECON'83,
Athens, Greece, May 1983, pp. A10. 02/1-4.
B. K. Bhattacharya, and G. T. Toussaint, Fast
algorithms for computing the diameter of a finite planar
set, The Visual Computer, vol. 3, no. 6, pp. 379-388,
1988.
D. Avis, G. T. Toussaint, and B. K. Bhattacharya, On
the multimodality of distances in convex polygons,
Computers and Mathematics with Applications, vol. 8,
no. 2, pp. 153-156, 1982.
K. Ichida, and T. Kiyono, Segmentation of plane
curves, Transactions of the Institute of Electronics and
Communication Engineers of Japan, vol. 58-D, pp. 689696, 1975.
G. T. Toussaint, On the complexity of approximating
polygonal curves in the plane, Proceedings of IASTED
International Symposium on Robotics and Automation,
1985, pp. 59-62.
G. T. Toussaint, Complexity, convexity, and
unimodality, International Journal of Computer and
Information Sciences, vol. 13, no. 3, pp. 197-217, 1984.
Y. Kurozumi, and W. A. Davis, Polygonal
approximation by the minimax method, Computer
Graphics and Image Processing, vol. 19, pp. 248-264,
1982.
H. Imai, and M. Iri, Polygonal approximations of a
curve: Formulations and solution algorithms, in
Computational Morphology, G. T. Toussaint, (Ed.),
Amsterdam, The Netherlands: North-Holland, 1988, pp.
71-96.
M. E. Houle, and G. T. Toussaint, Computing the
width of a set, IEEE Transactions Pattern Analysis &
Machine Intelligence, vol. 10, no. 5, pp. 761-765, 1988.
S. Drazi, N. Ralevi, and J. Zuni, Shape elongation
from optimal encasing rectangles, Computers and
Mathematics with Applications, vol. 60, pp. 2035-2042,
2010.
P. L. Rosin, Measuring shape: ellipticity,
rectangularity, and triangularity, Machine Vision and
Applications, vol. 14, pp. 172-184, 2003.
J. uni and P. L. Rosin, A new convexity measure for
polygons, IEEE Transactions on Pattern Analysis and
Machine Intelligence, vol. 26, no. 7, pp. 923-934, 2004.
14. J. uni and P. L. Rosin, Rectilinearity measurements

for polygons, IEEE Transactions on Pattern Analysis
and Machine Intelligence, vol. 25, no. 9, pp. 1193-1200,
2003.
15. J. uni and P. L. Rosin, and L. Kopanja, On the
orientability of shapes, IEEE Transactions on Image
Processing, vol. 15, no. 11, pp. 3478-3487, 2006.
16. H. Freeman and R. Shapira, Determining the
minimum-area encasing rectangle for an arbitrary closed
curve, Communications of the A.C.M., vol. 18, pp.
409-413, 1975.
17. F. C. A. Groen, P. W. Verbeek, N. de Jong, and J. W.
Klumper, The smallest box around a package, Pattern
Recognition, vol. 14, nos. 1-6, pp. 173-178, 1981.
18. G. T. Toussaint, Pattern recognition and geometrical
complexity, Proceedings of the Fifth International
Conference on Pattern Recognition, Miami Beach,
December 1980, pp. 1324-1347.
19. T. Needham, A visual explanation of Jensens
inequality, American Mathematical Monthly, vol. 100,
no. 8, pp. 145-185, 1993.
20. V. Klee, and M. C. Laskowski, Finding the smallest
triangles containing a given convex polygon, Journal
of Algorithms, vol. 6, no. 3, pp. 359-375, 1985.
21. J. ORourke, A. Aggarwal, S. Maddila, and M. Baldwin,
An optimal algorithm for finding minimal enclosing
triangles, Journal of Algorithms, vol. 7, pp. 258-269,
1986.
22. S. Das, P. P. Goswami, and C. S. Nandy, Smallest kpoint enclosing rectangle and square of arbitrary
orientation, Information Processing Letters, vol. 94,
pp. 259-266, 2005.
23. J. S. B. Mitchell, and V. Polishchuk, Minimumperimeter enclosures, Information Processing Letters,
vol. 107, pp. 120-124, 2008.
24. D. Mount, The densest double-lattice packing of a
convex polygon, J. E. Goodman, R. Pollack, W.
Steiger, (Eds.), DIMACS Series in Discrete
Mathematics and Theoretical Computer Science, vol. 6,
1999, pp. 245-262.
25. K. Hosono, H. Meijer, and D. Rappaport, On the
visibility graph of convex translates, Discrete Applied
Mathematics, vol. 113, pp. 195-210, 2001.
26. R. O. Duda, and P. E. Hart, Pattern Classification and
Scene Analysis, Wiley, New York, 1973.
27. B. K. Bhattacharya, and G. T. Toussaint, Efficient
algorithms for computing the maximum distance
between two finite planar sets, Journal of Algorithms,
vol. 14, pp. 121-136, 1983.
28. G. T. Toussaint, and J. A. McAlear, A simple O(n log
n) algorithm for finding the maximum distance between
two finite planar sets, Pattern Recognition Letters, vol.
1, pp. 21-24, 1982.
29. T. Lozano-Perez, An algorithm for planning collisionfree
paths
among
polyhedral
obstacles,
Communications of the ACM, vol. 22, pp. 560-570,
1979.
384
30. T. Lozano-Perez, Spatial planning: A conguration
space approach, IEEE Transactions on Computers, vol.
32, no. 2, pp. 108120, 1983.
31. R. L. Graham, An efficient algorithm for determining
the convex hull of a finite planar set, Information
Processing Letters, vol. 1, pp. 132-133, 1972.
32. F. P. Preparata, and S. Hong, Convex hulls of finite
sets of points in two and three dimensions,
Communications of the ACM, vol. 20, pp. 87-93, 1977.
33. B. Chazelle, Approximation and decomposition of
shapes, in Advances in Robotics 1: Algorithmic and
Geometric Aspects of Robotics, (J.T. Schwartz, C.K.
Yap, eds), Lawrence Erlbaum Associates (1987), pp.
145-185.
34. J. Abrahamson, A. Shokoufandeh, and P. Winter,
Euclidean TSP between two nested convex obstacles,
Information Processing Letters, vol. 95, pp. 370-375,
2005.
35. F. P. Preparata, and D. E. Muller, Finding the
intersection of n half-spaces in time O(n log n),
Theoretical Computer Science, vol. 8, no. 1, pp. 45-55,
1979.
36. J. ORourke, A new linear algorithm for intersecting
convex polygons, Computer Graphics and Image
Proccessing, vol. 19, pp. 384-391, 1982.
37. G. T. Toussaint, A simple linear algorithm for
intersecting convex polygons, The Visual Computer,
vol. 1, pp. 118-123, 1985.
38. Y. Freund, and R. E. Schapire, Large margin
classification using the perceptron algorithm, Machine
Learning, vol. 37, no. 3, pp. 277-296, 1999.
39. M. E. Houle, Algorithms for weak and wide separation
of sets, Discrete Applied Mathematics, vol. 45, pp.
139-159, 1993.
40. M. E. Mavroforakis, and S. Theodoridis, A geometric
approach to support vector machine (SVM)
classification, IEEE Transactions on Neural Networks,
vol. 17, no. 3, pp. 671-682, 2006.
41. C. Rey, and R. Ward, On determining the on-line
minimax linear fit to a discrete point set in the plane,
Information Processing Letters, vol. 24, pp. 97-101,
1987.
42. B. K. Bhattacharya, J. Czyzowicz, P. Egyed, I.
Stojmenovic, G. T. Toussaint, and J. Urrutia,
Computing shortest transversals of sets, International
Journal of Computational Geometry and Applications,
vol. 2, no. 4, pp. 417-436, 1992.
43. J.-M. Robert, and G. T. Toussaint, Linear
approximation of simple objects, Proceedings 9th
Annual Symposium on Theoretical Aspects of
Computer Science (STACS'92), Cachann France,
February 1992, A. Finkel and M. Jantzen (eds.), Lecture
Notes in Computer Science #577, pp. 233-244.
44. J.-M. Robert, and G. T. Toussaint, Computational
geometry and facility location, Proceedings of the
International Conference on Operations Research and
Management Science, Manila, The Philippines,
December 11-15, 1990, pp. B-1 to B-19.
45. B. Aronov, D. Garijo, Y. Nez-Rodrgues, C. Seara,

and J. Urrutia, Measuring the error of linear separators
on linearly inseparable data, Discrete Applied
Mathematics, vol. 160, 2012.
46. U. Grenander, Pattern Synthesis, Springer-Verlag, New
York, 1976.
47. G. Barequet, and B. Wolfers, Optimizing a strip
separating two polygons, Graphical Models and Image
Processing, vol. 60, no. 3, pp. 214-221, 1998.
48. P. Bose, F. Hurtado-Diaz, E. Omaa-Pulido, J.
Snoeyink, and G. T. Toussaint, Some aperture-angle
optimization problems, Algorithmica, vol. 33, pp. 411435, 2002.
49. M. Teichmann, Wedge Placement Optimization
Problems, M.Sc. Thesis, School of Computer Science,
McGill University, October 1989.
50. J. P. Rasson, and V. Granville, Geometrical tools in
classification, Journal of Computational Statistics and
Data Analysis, vol. 2, pp. 105-123, 1996.
51. G. T. Toussaint, New results in computational
geometry relevant to pattern recognition in practice, In
Pattern Recognition in Practice II, E. S. Gelsema and L.
N. Kanal, (Eds.), North-Holland, Amsterdam, The
Netherlands, 1986, pp. 135-146.
52. B. Chazelle, On the convex layers of a planar set,
IEEE Transactions on Information Theory, vol. 31, pp.
509-517, 1985.
53. G. T. Toussaint, Quadrangulations of planar sets,
Proceedings of the 4th International Workshop on
Algorithms and Data Structures (WADS'95), Kingston,
Canada, August 16-18, 1995, pp. 218-227.
54. J. Iwerks, and J. S. B. Mitchell, Spiral serpentine
polygonization of a planar point set, In Encuentros de
Geometria Computacional, (Hurtado Festschrift), A.
Mrquez, et al. (Eds.): LNCS 7579, 2012, pp. 146-154.
55. H. Brnnimann, L. Kettner, M. Pocchiola, and J.
Snoeyink, Counting and enumerating pointed pseudotriangulations with the greedy flip algorithm, In:
Algorithm Engineering and Experiments (ALENEX'05),
Vancouver, BC, Canada, January 2005.
56. D. Eppstein, Average case analysis of dynamic
geometric optimization, Computational Geometry:
Theory and Applications, vol. 6, pp. 45-68, 1996.
57. R. Fulek and J. Pach, A computational approach to
Conway's
thrackle
conjecture,
Computational
Geometry: Theory and Applications, vol. 44, pp. 345
355, 2011.
58. J. Pach and E. Sterling, Conway's conjecture for
monotone thrackles, American Mathematical Monthly,
vol. 118, no. 6, pp. 544548, 2011.
59. S. L. Devadoss, and J. ORourke, Discrete and
Computational Geometry, Princeton University Press,
Princeton, New Jersey, 2011.
60. F. P. Preparata, and S. J. Hong, Convex hulls of finite
sets of points in two and three dimensions,
Communications of the ACM, vol. 20, no. 2, pp. 8793,
1977.
385
61. J. O'Rourke, Finding minimal enclosing boxes,
International Journal of Computer and Information
Science, vol. 14, pp. 183-199, 1985.
62. C.-T. Chang, B. Gorissen, and S. Melchior, Fast
oriented bounding box optimization on the rotation
group SO(3;R), ACM Transactions on Graphics, vol.
30, no. 5, October 2011, Article 122.
63. G. Barequet, and S. Har-Peled, Efficiently
approximating the minimum volume bounding box of a
point set in three dimensions, Journal of Algorithms,
vol. 38, pp. 91-109, 2012.
64. I. T. Jolliffe, Principal Component Analysis. Springer,
New York, NY, USA, 2002.
65. M. Lahanas, T. Kemmerer, N. Milickovic, K.
Karouzakis, D. Baltas, and N. Zamboglou, Optimized
bounding boxes for three-dimensional treatment
planning in brachytherapy, Medical Physics, vol. 27,
no. 10, pp. 2333-2342, October 2000.
66. G. T. Toussaint, and B. K. Bhattacharya, Optimal
algorithms for computing the minimum distance
between two finite planar sets, Pattern Recognition
Letters, vol. 2, pp. 79-82, 1983.
67. C. Grima, and A. Mrquez, Computational Geometry on
Surfaces, Kluwer Academic Publishers, Dordrecht, The
Netherlands, 2001.
68. H. Pirzadeh, Computational Geometry with the Rotating
Calipers, M.Sc. Thesis, School of Computer Science,
McGill University, Montreal, Canada, November 1999.
386
Virtual Learning Environments: Proposals for Authoring and Visualization of

Educational Content
1
Tiago Cinto, 1Harlei M. A. Leite, 2Cecilia S. A. Peixoto and 1Dalton S. Arantes

School of Electrical and Computer Engineering, University of Campinas, UNICAMP
1
Av. Albert Einstein, 400, 13085-852 - Campinas, SP, Brazil
2
Salesian University of So Paulo UNISAL, Campinas, SP, Brazil
1
{tcinto, hmleite, dalton}@decom.fee.unicamp.br, 2cecilia.peixoto@sj.unisal.br
ABSTRACT
Heated debates involving reforms in the
educational system are becoming more and more
frequent in recent years. This is due to the
increasingly evident shortcomings in the
educational system and its difficulties to evolve at
the same pace as technological development. The
aim of this work is to present an analysis of the
current engineering educational context and
propose an interactive content authoring system as
well as a virtual professor in an attempt to make
learning experience richer and more motivating to
students.
KEYWORDS
Flipped Classroom; Virtual Learning Environment;
Avatar; E-Learning; Content Authoring.
1 INTRODUCTION
The advent of new technologies in recent years
has been transforming human lifestyle at a
breathtaking speed. Education, as an exception, is
one of the social structures most resistant to
adopting the technological tools available today.
Educators, in general, do not truly believe that the
lack of interest and low student performance can
be directly associated with the traditional and
restricted way of learning. This has motivated the
interest of researchers in the search for
methodologies and processes in an attempt to
make an educational reform based on Digital
Information and Communication Technologies
[1].
In applied sciences, for instance, there is a
growing disaffection of high school students for
courses in the STEM areas (Science, Mathematics,

Engineering and Technology). Students do not
consider these fields creative or even innovative
enough. This lack of interest is often due to
misconceptions on the role and responsibility of
professionals working in these areas. The
consequences of this scenario will certainly have
profound impact on industry development,
especially in emerging countries. In Brazil, for
instance, there is currently a shortage of 150,000
qualified engineers, according to the Brazilian
National Industry Confederation - CNI. The
proportion in the country is 6 engineers per 1,000
economically active people. In developed
countries, such as U.S. and Japan, the proportion
is 25 engineers per 1,000 workers [2].
The current generation of students has grown
interacting with computers, video-games and other
technological devices. This makes the paradigm of
learning-by-doing even more important today.
As a crucial complement to traditional methods,
this paradigm should not be overlooked in new
proposals for youth training. Everyone seems to
agree that the best way to learn is by putting into
practice the acquired knowledge in everyday
needs [3].
In this context, researchers have been
proposing methodologies and tools to overcome
the problems described above in an attempt to
make school more attractive for younger
generations. Marasco and Behjat [4], for instance,
have proposed a multidisciplinary framework for
electronic and electricity teaching in Electrical
Engineering courses. Their aim was also to
integrate the disciplines of English, Physics,
Sociology and Arts in the traditional curriculum.
This framework is based on the CDIO paradigm
(Conceive, Design, Implement and Operate) and
387
PBL (Problem-Based Learning). Other proposals
refer to the intensive use of WebLabs, Virtual
Labs and Simulation Tools, either at home or in
the classroom [5, 6, 7, 8, 9]. However, these new
methods of education will require: (1) improved
classroom acoustic and environments; (2) highquality audio and video; (3) use of smart
computers; (4) use of Flipped Classroom
methodology and (5) proposition of real-life
challenges.
Currently, with the help of Internet, millions of
students can be simultaneously attended in online
courses offered in different areas of knowledge.
Universities like Harvard, MIT and Stanford have
already adopted distance learning and achieved
good results with their MOOCs (Massive Online
Open Courses) offered through Coursera and edX
platforms.
In addition to distance learning, the concept of
Flipped Classroom [10] seems to be an efficient
solution to make classes more dynamic and
attractive. Reversing a classroom means to change
the pedagogical model of teaching concerning the
traditional oral exposition of concepts and
homework assignment. In this new model,
students are encouraged to study and learn by
themselves, reading books and articles, watching
videos and solving especial homeworks. The
classroom environment is then redirected to focus
on projects, activities and discussions. Technology
and new media are invaluable tools for the success
of this concept. Briefly, the value of this inversion
is the transformation of disciplines into
workshops, where students can test their
knowledge in the application of concepts and
interact with each other in practical activities [10].
In other words, learn at home and practice in the
classroom.
According to [10], in traditional classes
students try to understand what is being said,
which often makes them lose important points of
the speech. They also cannot think about the
speech often because they are trying to transcribe
the words of their instructors. The use of movie
and other technological resources places them
under control of the content presentation execution
flow. It is worth mentioning that there is also the
possibility to include people with special needs
through the use of subtitles and other interesting
features. By delegating instructional time to these

resources, professors are able to concentrate on
subtle aspects of learning, such as students'
inherent difficulties, misconceptions in reasoning
and interpretation, etc.
There is not a well-defined model when it
comes to Flipped Classrooms. However, it usually
contains pre-class contents being offered to
students in order to transform classes in
discussions, projects and activity solving.
According to [10] and [11], a simple and
traditional style to be applied consists of students
watching pre-class movie lessons up to 20
minutes, followed by quizzes and online activities
to measure how much they learned, which allows
them to have immediate feedback. In the
classroom, instructors are advised to stimulate
discussion groups or even turn the classroom into
practical laboratory for students to practice what
they have learned.
The use of Flipped Classrooms has become
recurrent in recent years, especially due to its
proven benefits [11, 12, 13, 14, 15]. Although
recent, to the extent that the concept becomes
popular, new proposals for its adaptation to make
use of new technological resources are created.
Particularly, with the emergence of smartphones,
tablets and laptops the access to this type of
education becomes ubiquitous and therefore can
be accessed from anywhere. It is also worth
emphasizing the need for Universities and
Institutions to be adapted to accommodate this
new format of pedagogy [10]. Among the main
benefits found by researchers we can list:
More content could be taught by
professors;
Students
performance
was
better
compared to traditional tests and exams at
the end of the course;
Although students felt uncomfortable at
first, they were able to quickly adapt
themselves to this format in order to
consider it satisfactory and effective.
Economides [16] included Flipped Classrooms
in his state-of-art analysis of the most valuable
educational technologies current being used these
days:
Massive Online Open Courses (MOOCs);
388
Flipped Classrooms;
Online video broadcasting (YouTube);
Learning Management Systems (LMS);
Real-time communication;
Increased use of technological devices
(smartphones, mobile phones, tablets, etc.);
Broadband Internet.
According to Economides [16], the term
Collaborative Education should be used instead
of Distance Education, considered by him
surpassed.
Although simple and promising, the reverse
classroom paradigm can be easily misunderstood.
This occurs mainly because of the careful
preparation of contents that are needed. Moreover,
lesson recording requires great dedication and
effort, and activities need to be carefully
integrated with content taught. In addition,
instructors may need to learn new skills to put into
practice lessons recording, which introduces a
learning curve for the model. Well-structured and
well-prepared lessons contribute to students'
motivation [10]. Not to mention the cost of
recording equipment and the difficulties of
updating or editing recorded content. Our proposal
tries to deal with these trade-offs and will be
further discussed soon.
Peer Instruction [17] has also been used in a
similar way. Proposed by Eric Mazur, it involves
students completing simple and direct learning
tasks at home and discussing content in classes
assisted by their instructors. This technique was
first used in introductory physics courses at
Harvard University and quickly spread to other
disciplines and institutions due to its efficiency
and flexibility compared to traditional teaching
methods based on oral presentation [18, 19, 20].
Watkins and Mazur [21] have found a significant
increase of retention among STEM students that
used this technique in an introductory physics
course. Giving students the chance to think,
respond and actively interact in the classroom can
significantly contribute to retention in courses in
these areas [21].
The Peer Instruction technique consists of:
1. Professor proposes a pre-class activity that
might be a reading, movie, etc., as
homework;
2. In class, professor proposes questions

based on student's feedback to pre-class
readings;
3. Students think about them in peers;
4. Students commit to a consensus;
5. Professor review answers given;
6. Students discuss new ideas and answers
with their peers;
7. Again, students commit to a consensus;
8. Again, professor reviews the answers and
decides where more explanation is needed
before changing to a new concept for the
cycle to be repeated.
In the 90s, Professor Eric Mazur [17]
suggested what would be the teaching paradigms
of current days, mainly transformed by the use of
computers and technological devices in the
learning process:
I believe that we are just seeing the
beginning of the process and the computer
will soon become an integral part of
education. Computers will not replace
teachers, but will surely provide dynamic
tools to improve the quality of education.
Along with Flipped Classrooms, Peer

Instruction and Simulation Tools, Avatar-based
proposals for teaching students have also been
published in recent years [22, 23, 24, 25]. It has
been recognized that one of the main problems
faced by these methodologies is the lack of
interest of students in spending time at home to
learn by themselves [8]. Our group is currently
working on attractive ways to present content to
students to be used along with these new
proposals. We are building two platforms: one in
which the instructor can build especial educational
contents and another in which students can watch
contents being presented by virtual instructors.
1.1 Related Work
The use of virtual tutors within computer aided
learning has become recurrent in research agendas
nowadays [22, 23, 24, 25, 26, 27].
Virtual tutors are characters able to entertain
his interlocutors while they learn. The work of
Grujic et al. [22] proposes a virtual tutor to be
used in a virtual educational system. The avatar is
389
integrated into an e-learning system and is able to
interact with users sympathetically. Its goal is not
actually to teach students, but demonstrate
emotional reactions according to actions taken by
them, with the use of gestures and facial
expressions.
According to Wang et al. [23], it is important
to make students to feel good during the learning
process. In this context, in their work is proposed
an approach to build an environment for empathic
learning. Users can express their emotions while
they learn. A virtual human was also created for
proactively reacting to users' actions.
Other proposals of virtual professors were
made by Hu and Zao [24] and Sharma et al. [25].
In their work is proposed a 3D virtual teaching
environment where students can connect and
interact remotely with each other and with their
professors, and attend classes, all of them
represented by avatars. The virtual professor is
suggested to be affective and intelligent when
interacting with students according to them.
1.2 Challenges for Contemporary Educators
Currently, we spend much of the day interacting
directly or indirectly with technological devices.
All this involvement can be used for educational
purposes. Through this interaction, we have easy
access to a large number of information, what was
difficult to access in the past.
With all this advancement, and considering
that most students have access to technological
devices, we can think about a methodology that
enhances teaching based on the conceptualization
of knowledge rather than just memorization.
However, in order to take advantage of the
benefits provided by technology, we must have a
teaching environment that makes it possible to
apply new methodologies. Hence, this involves
creating computational tools to improve education
and motivating student's interaction with
professors, monitors, and classmates.
To accommodate all technological capability it
is necessary to adapt classrooms, so that all
available resources can be better used. This means
providing classrooms with high definition
projectors, high quality sound systems, digital
whiteboards, tablets, notebooks, access to
broadband network, and so on. In addition to

encouraging students to use technology as a means
of study, providing means for everyone to have
their own equipment is also very important.
It is also important to encourage teachers to
create new methodologies that use all
technological resources available for classroom
improvement, besides creating teams of
professionals composed of experienced and
beginning teachers and monitors of various levels
(masters, doctoral and postdoctoral). These
heterogeneous teams are able to more effectively
provide theoretical and practical instructions
together in the same classroom, as well as to
provide efficient use of simulation tools for
visualization and interaction.
Massive amounts of information are freely
available today on the Internet. From simple
images to extensive free encyclopedias, the
content is immeasurable and invaluable. With the
simple click of a mouse and a few search
keywords, a great number of priceless information
content is presented. In this context, making use of
this giant supply of information is of a great value
for teachers.
Encouraging students to look for content on
the Internet, as well as teaching them techniques
for positive content filtering, contributes
significantly with the learning process by exposing
students to other universes beyond the textbooks
and handouts.
Motivating students to carry out individual and
team projects, as well as presenting them to their
classmates, allows them to actively and effectively
interact with the context of the discipline. In this
way, learners are encouraged to look for content
outside the class to understand the problem, and
often develop and work their own solutions, which
makes them feel the nice experience of being an
expert in the subject. Also, the presentation and
discussion of projects makes them familiar with
actual pedagogical situations, where they are
forced to play the role of tutors in front of the
room, even as if for a short period of time.
The simple stimulus for learning foreign
languages and methods of technical and scientific
wording considerably expands the horizon of
students' experience. This means they can be
familiar with information available in other
390
languages rather than just in their mother tongue.
In addition, students learn valuable techniques for
writing and reading academic papers.
2 PROPOSAL OF AVATAR-BASED
TEACHING
modeled by the Editor. In this platform we have

two avatars acting as virtual professors, a male
and a female. Briefly, the Player is composed of
chalkboards, projection screens, computers, desks
and books. It simulates a real classroom in which
the avatar interacts with the environment.
The purpose of this work is mainly based on the

creation of virtual classes taught by avatars in elearning environments. Here we propose an
environment where a human professor works by
creating classes that are taught by avatars and
watched by students.
Figure 2. Virtual Classroom.
Figure 1. Virtual Class Editor.
Currently, work is being done on two distinct

platforms: an editor of virtual classes and a player
for them. The Virtual Class Editor in Figure 1 is
intended to support professors designing virtual
classes in a friendly and simplified format. The
class authoring is made through timeline
paradigm, based on execution of temporal scenes.
Rich media can be inserted in scenes such as
movies, images, audios, documents, URLs, etc.
All the actions and gestures performed by the
virtual professor are controlled by the human
professor in this editor.
The Virtual Classroom, presented in Figure 2,
is intended to support the execution of classes
In addition to the content used for traditional

exposure, the Virtual Class Editor also allows the
modelling of interactive activities such as quizzes,
questionnaires and reference materials for
students.
Some of the main benefits of the Editor
compared to traditional movies involve:
Little expertise is required from professors
to create content;
A wide variety of content can be included
in virtual classes;
Professors can edit and re-edit classes at
any time with relativity little effort.
Traditional movies usually require the
content to be recorded over and over;
There is no need to worry about codec
conversions or resolution issues to adjust
content for the Player. The Editor takes
care of these processes;
Low cost is required to create virtual
classes. Movies usually require expensive
recording instruments and people to
support it;
Students feel motivated to use our
approach (as we will show soon);
It represents a good alternative to be used
in conjunction with Flipped Classroom or
Peer Instruction methodology, since
391
students seem more motivated to study at
home.
2.1 Virtual Class Editor
The Virtual Class Editor was designed to

support the work of professors in the creation of
instructional content presented within the Virtual
Classroom. Professors can model scenes which are
presented by avatars, create interactive activities
for students, insert reference materials, etc.
The tool has structured its development on the
.NET platform, using C# coding language.
Besides the flexibility provided by this
framework, it can be said that its adoption
occurred because Virtual Classroom has been
coded in Unity3D, also developed using .NET. No
DBMS (Database Management System) was used
for the storage of data, only notations for file
serialization. The editor also made use of a
multimedia encoder (ffmpeg2theora) in order to
allow the work with the most common media
formats currently available on the market.
Some of the main components built to create
the editor were:
Language Parsers: the tool uses XML to
serialize class content as well as language
description. Therefore, in order to an
application be able to read and write XML,
some parsers are necessary;
TTS System: this is necessary to allow the
creation of synthetic voice based on text
for the avatars. The tool does not make use
of voice recording;
Multimedia Player: the multimedia player,
as shown in Figure 3, is intended to allow
professors previewing multimedia content
when modeling classes. It works with
image, audio and movie files;
Multimedia Codec: this is necessary to
process a wide variety of media files as
input. These files are adjusted and encoded
to fit VClass Projects requirements as we
will see soon.
Process Builder: this component was built
to provide multi-threading support for the
tool. Several threads are used to optimize
processing of multimedia files;
Language Module: this element allows the

system language to be easily translated
throughout the translation of some
language files;
Cryptography
Module:
component
responsible for adding cryptography to
content created by the editor in order to
avoid unauthorized access to XML. A 256
bits security key was used along with the
Rijndael symmetric algorithm.
Figure 3. Virtual Class Editor Multimedia Player.
VClass Projects are structures for storing

virtual classes generated by the Virtual Class
Editor. It can be said that they correspond to a
well-defined organization hierarchy, composed of
files and folders used to organize content. We can
have as example the illustration in Figure 4.
VClass Projects have a nested directory named
Resource and a .vclass file. The Resource
container is the hub of resources, in other words, it
stores voice files, movies, pictures, audios, etc.
The .vclass file contains the virtual class
description language, specified using XML.
392
in particular. Therefore, we can say that some of
the non-functional requirements that guided the
development of this environment were:
Usability: to create friendly user interfaces
to assist professors and students;
Performance: to avoid excessive waiting in
multimedia processing;
Security: to prevent unauthorized access to
content of VClass Projects;
Integration: to allow Virtual Class Editor
to be integrated to any TTS (Text-toSpeech) system built using SAPI 5
framework from Microsoft;
Internationalization: to allow the editor to
be easily translated to any language
throughout the use of XML.
Figure 4. VClass Projects.
2.2 Virtual Classroom
The eXtensible Markup Language (XML) is a

recommendation created by the World Wide Web
Consortium (W3C), an international consortium of
companies and organizations created to develop
standards for processing of distributed Web
content. The use of XML allows the creation of
customizable markup languages to meet a special
need. Therefore, it can be said that the platform
has used this technique to describe its Virtual
Class Markup Language, also known as VCML.
Also according to the diagram shown in Figure
4, we can see some multimedia file formats such
as JPG, WAV and OGV. In a nutshell, they are the
formats adopted by the tool for recording and
viewing content. Images are stored in JPG and
movies in the open format OGV. Audios and
voice files are encoded to WAV.
At this point it is worth saying that for the
correct operation of both platforms, this hierarchy
must be maintained: VClass Projects that do not
follow this standardization are unable to be edited
by the editor or even be watched by students in the
Virtual Classroom.
Non-functional requirements of a system are
composed of restrictions relating to its services
and functions. They include performance
constraints, standards, database, security, etc.
These requirements are usually associated with the
system as a whole and not with some functionality
The Virtual Classroom was designed to

support students when viewing content created by
the Virtual Class Editor.
This tool made use of .NET development
platform, also using C# coding language.
MonoDevelop was used as the integrated
development environment (IDE). This application
is a suite of free compilers created as an
alternative to Visual Studio, supplied by
Microsoft. The C# language was a slightly adapted
version for programming in script mode. This
development paradigm is used in most 3D model
programming environments available today. This
platform also made use of Unity3D as rendering
engine for graphics processing.
Due to the characteristics of APIs and
frameworks adopted for the development of tools
of this work, it can be said that both are restricted
to Windows operating system and cannot be
considered cross-platforms.
The Virtual Classroom also used software for
modeling and graphical animation to work with
3D models of objects and avatars such as
Autodesk Maya and MotionBuilder, and Blender.
In 3D interactive virtual environments its
common to have the concept of HUDs (Head-Up
Displays). This concept refers to a graphical
interface that remains active in the screen during
the whole time that the application is executed.
393
Therefore, a HUD was modeled for the Virtual
Classroom in order to allow students to have
access to all the tools offered by virtual classes, as
shown in Figure 5. Regarding this picture, we can
also see several numbers indicating some specific
regions of this component:
1. Button that give us access to the playlist of
scenes;
2. Scenes control area. When any option of
this area is accessed, scenes can be
forwarded, rewound, paused, etc.;
3. Button that give us the ability to move to a
specific scene in the playlist;
4. When the student wants to access the
interactive quiz, this button can be used.
Quizzes may or may not be present, since
it depends on the professor to model them;
5. Button to access the class questionnaire;
6. Button that give us access to extra-class
resource materials. A wide range of items
can be added to classes as resource
materials to support students that want to
have further information about the topics
covered;
7. Button that provides access to the got
question interface. When the student clicks
on it, the class execution flow is
interrupted so he or she can write its doubt.
After this doubt is transmitted, the flow is
resumed;
8. Option that allows returning to the Virtual
Classroom main menu, composed mainly
of classes contained in the Workspace
directory;
9. The ability to show subtitles was coded in
the tool in order to support students when
hearing synthetic voices or even for
internationalization purposes. Subtitles can
be activated by this button;
10. The immediate interruption of any song
played by the tool can be made through
this option at any time;
11. And finally, the click on button 11 turns
off the execution of classes, as well as the
tool closed.
Figure 5. Virtual Classroom Head-Up Display.
According to Figure 5, its possible to see a

multimedia content being presented by the
projector in the class projection screen, in this case
a slide. We can also see that the avatar does not
participate in this moment. This strategy tries to
reduce the burden of visual elements shown
simultaneously to the student, in other words, in
order not to cause any distraction to the
explanation, only the relevant content is shown.
Therefore, the student hears only the voice of the
avatar on projected images, movies and audios.
The main menu interface is shown in Figure 2.
The content of this component is populated with
information from VClass Projects contained in the
Workspace folder. Projects outside this directory
are not displayed correctly and therefore cannot be
watched by students. This strategy eases the
distribution of classes by professors to their
students by the concentration of content in only
one place.
Also according to Figure 2, we can see the
avatars modeled to be presenters of content. For
animating these avatars models, human
movements were captured using Kinect, from
Microsoft. These movements could then be
segmented and populated in a base of gestures,
which served to support the creation of a Markov
chain to control the flow of execution of
movements.
In the interface shown in Figure 2 we can also
see some required fields that are asked to be filled
before the execution of virtual classes (for
instance, name, email and student id). This is
necessary for reporting purposes of students
scores in interactive activities, since these fields
394
do not constitute an access control system. Also
according to this picture, you may also notice an
option to display or omit scenes containing
content considered advanced. Marking content as
complex depends on the professors when they are
modeling scenes. This functionality was
implemented to meet a requirement generated
during evaluations, when students reported that
they were not willing to see content that did not
match their cognitive load. Therefore, now it is
possible to accommodate a wide diversity of
students' profiles.
When the Virtual Classroom is started, some
checkup operations are executed in order to
validate the consistency of the tool and projects
contained in the Workspace directory. If there are
any unwanted issues, the environment interrupts
its execution and notifies the user the problem
occurred.
Figure 6. Virtual Class Exhibition.
Figure 6 shows the presentation of a scene

being made by the female avatar. This
presentation is made by the movements
commented in previous paragraphs. In this picture
we can also see the use of subtitles synchronized
to the audio narration, besides the male avatar that
does not participate in scene waiting next to the
printer. For this case, there isnt effective use of
the chalkboard environment. The other items that
compose the classroom scenario (for instance,
books, computers, tables, etc.) are also there;
however, they do not represent interactive
elements for the avatar.
3 EVALUATION METHODOLOGY
In order to evaluate the attributes of the Virtual 3D
Learning Environment, we have selected especial
classes of graduate students in the School of
Electrical and Computer Engineering, University
of Campinas - UNICAMP, and in the Salesian
University of So Paulo UNISAL, both in the
city of Campinas, SP, Brazil.
A survey was carried out during the month of
November of 2013, where 73 individuals had the
opportunity to interact with classes presented by
avatars in our Learning Environment. At the end
of each presentation the students were asked to
answer an especial questionnaire.
The nature of the survey was based on mixedmethods. Data were collected in questionnaires
containing 12 Likert items [28], which are based
on a scale of levels of agreement or disagreement
to specific questions: (1) Completely Disagree, (2)
Disagree, (3) Indifferent, (4) Agree, (5) Strongly
Agree; and three open items.
The questionnaires were based on a proposal
by Knay and Knaack [29], which was developed
mainly to measure user satisfaction of students
when interacting with learning objects. We have
adapted their proposal for our study by adding
three open items. The original proposal divides the
instrument into 3 dimensions to group and collect
data:
Learning
1. The platform helped me learn.
2. Feedback resources (i.e., email) can be
useful in learning.
3. Graphical resources and animations helped
me learn.
4. The platform helped me learn a new
concept.
5. Overall, the learning platform helped me
learn.
Quality
6. Help features can be useful in learning.
7. Instructions appear to be easy to follow.
8. The platform appears to be easy to be used.
9. The platform appears to be organized.
Motivation
10. I liked the theme presented.
11. I consider the learning environment
motivating.
395
12. I would like to use the platform again.
Open Items
1. What did you like most in the platform?
2. What did you most dislike in the platform?
3. What would you change in the platform?
The questionnaire reliability analysis was
performed in accordance with the principles of
Cronbach's alpha coefficient [29], which ended up
showing that the questionnaire was reliable to be
applied in the study. Part of the instrument was
already validated in previous studies and presented
good results [30].
Cronbachs alpha coefficient can be estimated
according to
(1
(1)
where is the coefficient, is the amount of

items, is the variance of each item and is
the variance of all items.
Values greater than 0.7 for the coefficient are
said to be good and satisfactory. Values below this
limit are said to be poor and therefore can be
discarded [29].
For the analysis of quantitative data,
parametric studies concerning descriptive statistics
were chosen (average calculation, standard
deviation, etc.). For qualitative data, students
comments were categorized and scored in order to
enable their representation and visualization in
usual graphical forms [30]. The realization of a
pilot study to fit the standard deviation of the
sample was also contemplated.
All evaluation sessions were carried out in
groups, with the projection of virtual classes in a
multimedia projector for a classroom of students.
The purpose of the study as well as the main
features and capabilities of the environment were
also explained. At the end of the exposition
participants were asked to answer the
questionnaire. High quality loudspeakers were
used to improve sound quality, given the large
number of participants in each class.
In the next section we discuss the pilot study
carried out to obtain a reliable estimate for the
sample size of the population.
3.1 Pilot Study

Working with the entire population in a scientific
study is often impractical. Therefore, we have to
choose a representative sample of this population,
consisting of a typical subset of individuals.
The determination of the population sample
size is an issue of extreme importance, mainly for
two reasons:
Small sample sizes tend to lead to
inaccurate results, which makes the
extraction of reliable results impractical;
Large sample sizes tend to lead to waste of
resources and time employed in the survey.
It is possible to obtain a reliable estimate of
the population sample size by using
= (
( )
where,
= the number of individuals;
= the confidence interval (CI) [31];
= standard deviation of the sample;

= the error rate.
For a first approach of calculus, standard
deviation is usually not known. Therefore, many
authors suggest a pilot study with the largest
number of participants. After carrying out this
study, standard deviation can be extracted and
analyzed to estimate the sample size.
For this work we decided to carry out the pilot
study with the largest number of individuals.
Therefore, 47 graduate and undergraduate students
were chosen initially. Worst case was always
considered to estimate the size of our sample. For
this initial study, standard deviations were similar,
taking the mean as reference. Normalizing, the
standard deviation was about 20% for all items.
Therefore, for CI = 90% the sample size would be
44, for CI = 95% would be 62 and for CI = 99%
would be 106 people. For all these confidence
intervals a 5% survey error rate was considered.
Therefore, we have chosen to work with a
reliability of 95% since the number of individuals
selected for the final study was 73.
396
3.2 Sample Characteristics
A high percentage of individuals were enrolled in
an undergraduate course (89.04%, N = 66) while
others were involved with a graduate course
(9.59%, N = 7)
Except for one course, the others had
somewhat similar numbers of participants in this
study: 35.62% students were enrolled in Bachelors
in Mechanic Engineering (N = 26), 27.4% were
attending Industrial Automation (N = 20), 26.03%
were in Bachelors in Computer Engineering (N =
19) and, finally, 9.59% were enrolled in an
Electrical Engineering graduate course (N = 7).
Sample was mainly composed of young
people: 32.84% of respondents reported to be
between 21 and 25 years old (N = 22), 26.87% of
them said they were between 15 and 20 years old
(N = 18), 13.43% were between 26 and 30 (N =
9), and, finally, the rest of them declared to be in
age groups with smaller percentages.
The vast majority of respondents declared to
be males (91.67%, N = 66) while there were only
8.33% females for this study (N = 6).
4 RESULTS
As a case study, we have modeled some virtual
classes to be presented to students. Due to our
sample characteristics (most of them were enrolled
in an engineering course), we decided to create
some introductory classes concerning the use of
Matlab applied to Calculus. We tried to enrich the
platform as much as possible, using almost all
features and characteristics developed. Students
had the opportunity to watch content, ask
questions to our avatars, solve interactive quizzes,
access complimentary resources such as
documents and websites, etc. Special attention was
given to the TTS (Text-To-Speech) system in
order to produce voices for avatars as close as
possible to human voices, reducing pronunciation
errors or misspellings.
According to the statistical parameters of our
results, presented in Table 1, the mean was similar
for all items, as well as the standard deviations.
Mode and median parameters were also identical.
Table 1. Statistical Parameters of the Study.

Item
Me
Mo
Item
4
4
3.7
0.4
0.6
7
4
4
4.0
0.2
0.5
8
4
4
3.8
0.4
0.6
9
4
4
3.7
0.4
0.6
10
4
4
3.8
0.3
0.5
11
4
4
4.1
0.3
0.5
12
Me
Mo
4
4
4.0
0.5
0.7
4
4
4.1
0.5
0.7
4
4
4.1
0.4
0.6
4
4
4.0
0.4
0.6
4
4
3.8
0.4
0.7
4
4
3.8
0.3
0.5
Data from Cronbachs alpha coefficient

analysis in Table 2 confirmed that the
questionnaire was reliable to be used in our study.
All dimensions of the instrument achieved values
greater than 0.70, which corresponds to
satisfactory results [29].
Table 2. Description of the Questionnaire.
Scale
LOES-S
Learning
Quality
Motivation
Items
12
5
4
3
Possible Range
12-60
5-25
4-20
3-15
Internal Reliability
0.82
0.78
0.73
0.77
Overall, analyzing the results of the three

dimensions
originally
proposed
by the
questionnaire, all of them obtained satisfactory
results.
Our platform has achieved good results
concerning learning dimension according to
students opinions, as shown in Figure 7.
Although most of them agreed that they were able
to learn new concepts when watching virtual
classes, a considerable number of students rated to
be indifferent. In item 4, for instance, more than
30% neither agreed, nor disagreed. For this
scenario, it is worth emphasizing the need for
further studies to measure how much they actually
learned. Since this evaluation was based mainly on
summative aspects concerning user satisfaction, it
would be interesting to carry out formative
evaluations to identify real knowledge retention.
397
Quality items of our questionnaire were those
that achieved the highest scores, as shown in
Figure 8. All items scored higher than 80%.
Learning
100%
Indifferent
80%
Agree;
Strongly
Agree
60%
40%
Disagree;
Completely
Disagree
20%
0%
i1
i2
i3
i4
i5
70% of students considered the learning

environment motivating and almost 78% agreed
that they would watch virtual classes again if they
had the opportunity to do so. It is worth
emphasizing that the quality of virtual classes
depends mainly on the content created in the
Editor. Well-designed classes tend to contribute to
motivation, while poorly-designed ones lead to
students rejection. It became clear that wellprepared classes were successful in trying to make
learning more attractive to students.
Motivation
100%
80%
Figure 7. Learning Results Chart.
60%
Although named quality, items in this

dimensions related mainly to help features or the
ease of use of the platform. These results reflect
mainly our initial requirements concerning
usability and functionality. Both platforms always
kept in mind to be as simples as possible in order
to be used by anyone.
40%
Quality
100%
Indifferent
80%
60%
Agree;
Strongly
Agree
40%
20%
Disagree;
Completely
Disagree
0%
i6
i7
i8
Indifferent
i9
Figure 8. Quality Results Chart.
The results of motivation dimension of the

questionnaire, as in Figure 9, show that students
felt motivated when they had the opportunity to
watch our classes. About 80% of them liked the
overall theme presented, focused on an
introduction to Calculus using Matlab. More than
Agree;
Strongly
Agree
20%
Disagree;
Completely
Disagree
0%
i10
i11
i12
Figure 9. Motivation Results Chart.
For qualitative data, students comments from

open items were grouped in some thematic
groups: theme, audio, ease of use, interaction,
animations, graphs, comparison to other methods,
motivation, technology, visual, challenge and
learning. After this grouping, each comment was
rated individually in a scale spanning -2 (very
negative), -1 (negative), 0 (neutral), 1 (positive)
and 2 (very positive). This grouping and scoring
allowed us to analyze qualitative data through
descriptive analysis.
Among main pros pointed out by students we
can list:
Similarity with a real classroom;
The motivation provided by the platform,
which is not based only on readings and
enables the interaction with a virtual tutor;
The fact that the environment is fun and
versatile.
As points of improvements, the suggestions
concerned fixing some visual aspects, such as
398
image resolution and fonts, and reducing speech
speed in order to improve understanding of what
was being spoken by the virtual tutors.
6.
5 CONCLUSIONS
Throughout this work we presented in-progress
studies of an attempt to create an environment to
help professors designing content-rich interactive
virtual classes, as well as virtual instructors.
Efforts have been made in the design of robust and
easy-to-use platforms for use in educational
contexts in order to enable the creation of
motivating and friendly classes. Our platform has
already achieved the first results. Summative
evaluations were carried out with success,
resulting in good feedback from users
satisfaction. On the other hand, we intend to carry
out formative evaluations to measure the extent to
which students have really learned from content
presentation. Moreover, there are several recent
developments with great potential for important
contributions to education. Big Data algorithms
and Artificial Intelligence, for instance, are among
the applications that are likely to foster the
advancement of these educational tools, especially
for Massive Online Open Courses.
7.
8.
9.
10.
11.
12.
13.
6 ACKNOWLEDGEMENTS
14.
The authors would like to thank So Paulo

Research Foundation - FAPESP, Padtec S.A.
(Proc. 2007/56018-4), CAPES and CNPq for the
financial support.
7 REFERENCES
1.
2.
3.
4.
5.
Almeida, M. E. B., Valente, J.A.: Integrao Currculo e

Tecnologias e a Produo de Narrativas Digitais. Currculo
Sem Fronteiras 12(3), 57--82 (2012)
Nascimento, C.: Faltam 150 mil Engenheiros no Pas,
http://economia.estadao.com.br/noticias/economia,faltam150-mil-engenheiros-no-pais,131609,0.htm
Papert, S., Harel, I.: Situating Constructionism. Ablex
Publishing Corporation, Norwood, USA (1991)
Marasco, E., Behjat, L.: Integrating Creativity Into
Elementary Electrical Engineering Education Using CDIO
and Project-Based Learning. In: Proceedings of the IEEE
International Conference on Microelectronic Systems
Education (MSE), pp. 44--47, Austin, USA (2013)
Apse-Apsitis, A., Avotins, A., Krievs, O., Ribickis, L.:
Pratically Oriented e-Learning Workshop for Knowledge
15.
16.
17.
18.
19.
20.
Improvement in Engineering Education. In: Proceedings of

the Global Engineering Education Conference (EDUCON),
pp. 1--5, Marrakech, Morocco (2012)
Casini, M., Member, S., Prattichizzo, D., Vicino, A.: The
Automatic Control Telelab: A User-Friendly Interface for
Distance Learning. IEEE Transactions on Education 46(2)
252--257 (2003)
Carreras, F., Snider, A.: A MATLAB Educational
Software Tool for Teaching Complex Analysis
Engineering. In: Proceedings of the IEEE Southeastcon
Engineering for a New Era, pp. 204--207, Orlando, USA
(1998)
Moreira, V. R., Cardoso, F. A. C. M, Arantes, D. S.:
Plataforma Reconfigurvel para Ensino e Pesquisa em
Laboratrio de Sistemas Digitais a Distncia. In:
Proceedings of XIX Simpsio Brasileiro de Informtica na
Educao, Fortaleza, Brazil (2008)
Moreira, V. R., Cinto, T., Leite, H. M. A., Arantes, D. S.:
Aprimorando o Ensino de Engenharia com Novas
Abordagens Usando Recursos Computacionais. In:
Proceedings of VI Congresso Tecnolgico InfoBrasil TI &
Telecom, Fortaleza, Brazil (2013)
Educause: 7 Things You Should Know About Flipped
Classrooms. ELI 7 Things You Should Know, 2012.
https://net.educause.edu/ir/library/pdf/ELI7081.pdf
Zappe, S., Leicht, R., Messner, J., Litzinger, T., Lee, H.W.:
Flipping the Classroom to Explore Active Learning in a
Large Undergraduate Course, American Society for
Engineering Education (2009)
Mason, G. S., Shuman, T. R., Cook, K. E.: Comparing the
Effectiveness of an Inverted Classroom to a Traditional
Classroom in an Upper-Division Engineering Course.
IEEE Transactions on Education 56(4), 430--435 (2013)
Amresh, A., Carberry, A. R., Femiani, J.: Evaluating the
Effectiveness of Flipped Classrooms for Teaching CS1. In:
Proceedings of the IEEE Frontiers in Education (FIE), pp.
733--735, Oklahoma, USA (2013)
Herold, M., Lynch, T., Ramnath, R., Ramanathan, J.:
Student and Instructor Experiences in the Inverted
Classroom. In: Proceedings of IEEE Frontiers in Education
(FIE), pp. 1--6, Seattle, USA (2012)
Strayer, J. F.: How Learning in an Oriented Classroom
Influences Cooperation, Innovation and Task Orientation.
Learning Environments Research 15(2), 171--193 (2012)
Economides, T.: The State of Art in Educational
Technology. In: Proceedings of the IEEE Global
Humanitarian Technology Conference, pp. 285--287, San
Jose, USA (2013)
Mazur, E.: Peer Instruction: A Users Manual. Upper
Saddle River, New Jersey, USA (1997)
Crouch, C., Mazur, E.: Peer Instruction: Ten Years of
Experience and Results. American Journal of Physics
69(9), 970 (2001)
Fagen, A., Crouch, C., Mazur, E.: Peer Instruction: Results
from a Range of Classrooms. The Physics Teacher (40),
206--209 (2002).
Rosenberg, J., Lorenzo, M., Mazur, E.: Peer Instruction:
Making Science Engaging. In: J. Mintzes (eds.) Handbook
of College Science Teaching, pp. 77--85, National Science
Teachers Association (2006)
399
21. Watkins, J., Mazur, E.: Retaining Students in Science,
Technology, Engineering, and Mathematics (STEM)
Majors. Journal of College Science Teaching 42(5), 36
(2013)
22. Grujic, Z., Kovacic, B., Pandzic, I. S.: Building Victor - A
Virtual Affective Tutor. In: Proceedings of 10th
International
Conference
on
Telecommunications
(ConTEL), pp. 185--189, Zagreb, Croatia (2009)
23. Wang, C., Chen, G., Liu, C., Liu, B.: Design an Empathic
Virtual Human to Encourage and Persuade Learners in eLearning Systems. In: Proceedings of the First ACM
International Workshop on Multimedia Technologies for
Distance Learning, pp. 27--32, New York, USA (2009)
24. Hu, Y., Zhao, G.: Virtual Classroom with Intelligent
Virtual Tutor. In: Proceedings of International Conference
on e-Education, e-Business, e-Management and eLearning, pp. 34--38, Washington, USA (2010)
25. Sharma, S., Agada, R., Ruffin, J.: Virtual Reality
Classroom as an Constructivist Approach. In: Proceedings
of the IEEE Southeastcon, pp. 1--5, Jacksonville, USA
(2013)
26. Amaral, W. M., Martino, J. M., Angare, L. M. G. Sign
Language 3D Virtual Agent. In: Proceedings of the 5th
International Multi-Conference on Society, Cybernetics
and Informatics (IMSCI), pp. 93--97 (2011)
27. Amaral, W. M., Martino, J. M.: Formalismo,
Implementao e Avaliao de um Sistema de Transcrio
para Gerar Contedo em Lngua de Sinais em Ambiente
Virtual. In: Proceedings of the V Encontro dos Alunos e
Docentes do Departamento de Engenharia de Computao
e Automao Industrial (EADCA) (2012)
28. Creswell, J. Research Methodology: Qualitative,
Quantitative, and Mixed Methods Approaches, 2nd ed.
SAGE Publications, Inc., Los Angeles, USA (2002)
29. Cronbach, L. J.: Coefficient Alpha and the Internal
Structure of Tests. Psychometrika 16(3), 297--334 (1951)
30. Kay, R., Knaack, L.: Assessing Learning, Quality, and
Engagement in Learning Objects: The Learning Object
Evaluation Scale for Students (LOES-S). Educational
Technology Research and Development 57(2), 147--168
(2008)
31. Triola, M.: Elementary Statistics. Pearson Education Inc.,
Boston (2006)
400
Miniaturized Bandstop Filters Using Slotted-Complementary Resonators

Mohammed M. Bait-Suwailam
Department of Electrical and Computer Engineering, College of Engineering
Sultan Qaboos University,
Muscat, Oman
E-mail: msuwailem@squ.edu.om
ABSTRACT
This paper presents a technique to miniaturize
bandstop
filters
applicable
for
radiofrequency/microwave applications. The technique
comprises the use of small resonant inclusions bridged
with a slot, etched from the metallic ground plane
underneath the filters microstrip line segment. Unlike
artificial magnetic materials, for instance split-ring
resonators (SRRs) that respond to vertical magnetic
field, complementary SRRs (CSRRs) resonate upon an
excitation of an axial time-varying electric field.
Numerical full-wave studies are presented here to
validate and proof the proposed concepts. It is found
that high miniaturization factors can be achieved for
the filters loaded with the proposed slotted
complementary resonators in comparison to nonslotted resonators. Furthermore, sensitivity analysis is
conducted to study the effect of geometrical parameters
of the proposed slotted resonators as well as the effect
of the choice of host media on the suppression
bandwidth and rejection level of such filters.
Miniaturized bandstop filters based on developed
slotted-complementary spiral resonators are also
proposed and numerically evaluated.
KEYWORDS
Bandstop filters, complementary-split ring resonators
(CSRRs), slotted-CSRRs, miniaturized filters.
1 Introduction
Microwave filters are vital building blocks in
modern wireless communication systems. Such
filters are widely used to either pass particular
frequency components or reject spurious modes or
harmonics. In particular, microwave bandstop
filters are used quite often in cellular base stations,
navigation systems and alike. Amongst the
available
bandstop
filters
manufacturing
technologies, the Microstrip line bandstop filters

(BSFs) are indeed more attractive in modern
Fig. 1 A split-ring resonator (SRR) particle with its relevant

dimensions. Note that gray area represents metallization.
wireless communication systems due to their low

cost and ease of integration with other radio
frequency/microwave circuits.
There exists variety of techniques to design
microstrip BSFs. One of which is to employ shunt
open-circuited resonators that are quarterwavelength long, which fits more to filter out
narrowband signals [1]. However, other
techniques, like the use of periodic structures,
defected ground plane structure are more desirable
for wideband communication systems [2]-[5].
Recently, subwavelength resonators, well-known
in literature as metamaterials, have received much
attention, especially for the synthesis of filters
design and harmonics suppression [6]-[8]. For
instance, split-ring resonator (SRR) that was
originally proposed in 1999 has widely been used
to artificially synthesize magnetic materials [9].
Pendry et. al in his pioneering work [9] proposed
the use of two concentric small metallic rings with
opposite cuts (see Fig. 1) to further enhance the
magnetic resonance of such rings. By cascading
401
enough periods of aforementioned concentric
rings and formed in either periodic or aperiodic
fashion, artificial materials can be realized with
unique effective magnetic permeability response.
When excited with an axial time-varying magnetic
field H (see Fig. 1), SRRs resonate and hence give
rise to an effective magnetic response, eff. It was
until that Smith and his ground [10] made the first
artificial magnetic material following the
hypothetical analysis of Veselago [11] and Pendry
to realize materials with unique effective electric
and magnetic properties.
It is important to highlight here that at frequencies
below the resonance frequency of the SRR
inclusion, the effective magnetic permeability
becomes positive, while at frequencies above the
SRRs resonance, eff becomes negative. On the
other hand, complementary split-ring resonator
(CSRR) is the dual-counter part of SRR [12]. In
other words, CSRR starts to resonate once excited
with an axial time-varying electric field E. Based
on Babenits principle [13], this means that
CSRRs give rise to an effective electric response,
eff. CSRRs are in fact achieved by replacing the
metallic rings of SRRs with apertures (i.e., slotted
rings) and surrounding free-space region nearby
SRR inclusion is replaced with metallic plate.
Fig. 2 (a) depicts a unit cell of CSRR inclusion.
For simplicity, a square shaped resonator is
shown. At resonance, the inductance from the
inclusions metals balances the capacitance
between the etched rings of the resonator.
Recently, slotted-CSRR inclusions were proposed
to mitigate space wave coupling effect between
microstrip patch antennas [14]. In Fig. 2 (b), a
modified resonator is proposed through the use of
two collinear CSRRs connected through a
compact slit. The slit (or bridging slot) provides
significant advantage in enhancing the bandwidth
of the proposed stopband filter.
This paper aims to explore the application of such
slotted resonators as particles to enhance
performance of bandstop filters. While in [15] the
aforementioned
application
was
briefly
highlighted, herein, more analyses are devoted to
thoroughly investigate the effect of geometrical
parameters as well as the applied host medium of
the proposed slotted complementary resonators on
the suppression band of the filters. Furthermore,

techniques to further miniaturize bandstop filters
are proposed, and numerically assessed.
Fig. 2 (a) CSRR unit cell, and (b) the proposed slottedCSRR inclusion. Note that gray area represents
metallization.
2 Design of bandstop filters based on Slotted

Complementary Resonators
Both SRR and its dual counterpart (CSRR) (see
Fig.2(a)) have widely been applied as building
blocks in variety of engineering applicants,
including filtering and harmonics rejection [16]. In
fact, the resonance frequency of either SRR or
CSRR inclusions can be estimated analytically
using quasi-static electromagnetic theory [6] or
alternatively
numerically
using
full-wave
characterization methods. At resonance, SRR or
CSRR behaves as an LC resonant tank circuit. In
this paper, more emphasis is devoted to the
numerical modeling aspects of bandstop filters
loaded with CSRRs and the proposed slotted
resonators. Two versions of slotted resonators are
proposed and studied here, namely: slotted-CSRRs
and slotted-spiral resonators (S-SRs).
A. Slotted-CSRRs based filters
The type of filter (S-CSRRs) is based on a
modified version of CSRRs. The new resonator
consists of two CSRRs bridged with a slot of
length Ls and width Ws as shown in Fig. 2(b). For
simplicity, a square shaped resonator is adopted. A
wide stopband rejection of signals is achieved by
402
linearly cascading the suppression band of both
CSRRs and the slot. By cascading several varied
in size S-CSRRs, a wide stopband behavior is
attainable. Fig. 3 shows the bandstop filter
comprises a transmission line segment of width
2.88 mm for an impedance of 50 . The proposed
resonator (slotted-CSRR) has the following
geometrical parameters (see Fig. 1): L = 4 mm, a =
b = g = 0.2 mm, Ls = 2 mm, and Ws = 0.35 mm,
where a dielectric substrate (r = 3.48, tan =
0.004) with a thickness of 1.27 mm is used. In this
study, more than one S-CSRR unit cell is used.
This helps to provide strong suppression (zero
transmission) when capturing the transmission
coefficient, S21, between two ports of a
transmission line segment as indicated in Fig. 3.
The periodicity (separation distance between each
successive S-CSRR inclusion) is maintained
constant at 1 mm for ease of manufacturing.
made to classical CSRR cell. In order to keep the

resonance of CSRR lies within the S-CSRR cell, a
bigger CSRR cell is therefore needed. For
instance, to achieve bandstop suppression at
around 5 GHz, the CSRR unit cell would require
an outer area of 4.5 x 4.5 mm2. Other
complementary configurations are also shown for
comparison purposes. As can be seen from Fig. 4,
a suppression bandwidth (S21 < -15 dB) of more
than 1 GHz resulted when using 3 cells of SCSRRs. Furthermore, a wider suppression band
can be achieved by increasing the number of SCSRR cells. The classical CSRRs have
narrowband suppression in comparison to the
proposed filter. The use of the S-CSRR without
the slits resulted in narrowband suppression as
expected.
Fig. 4 Simulated transmission coefficient, S21, for the

bandstop filter with the proposed slotted-CSRRs, and
compared to a bandstop filter with classical CSRR
inclusions.
Fig. 3 (a) 3D view of the bandstop filter with SCSRRs

etched from ground plane; (b) bottom view of the
bandstop filter with slotted-CSRRs etched from the
ground plane.
To demonstrate the effectiveness of the slot of the

S-CSRR cell in the proposed filter, comparison is
A parametric study of the effect of the slot on the

filter suppression capability is shown next. The
slot width, Ws, is varied and the transmission
coefficient of the bandstop filter is numerically
computed. Fig. 5 depicts the transmission
coefficient for the filter when varying the width of
the slot. It can be inferred from Fig. 5 that a
decrease in the slot width, Ws, increases the
capacitance of the slit and hence causes the
403
rejection band of the filter to shift to lower
frequencies.
The slotted-spiral resonators follows similar

structural feature as that S-CSRRs. It is
understood that the capacitance of a metallic spiral
resonator is almost fourth of that of a split-ring
resonator. In other words, the capacitance of
complementary spiral resonator (CSR) is almost
twice that capacitance of the CSRR. Thus, more
miniaturization is achieved with the use of spiral
resonant inclusions rather than the SRRs. In this
paper, slotted-spirals, as will be shown later, are
indeed good candidates to provide size reduction
of the bandstop filters.
To better understand the physical behavior of
slotted-CSR, bandstop filters based on CSRs are
studied first. Fig. 7 depicts CSR unit cell and the
proposed (S-CSR) inclusion.
Fig. 5 A parametric study illustrates the effect of the SCSRR slot width, Ws, on the rejection band of the filter.
The attenuation coefficient for the proposed filter

with S-CSRRs is numerically computed and
compared with CSRRs (without the slit). As can
be seen from Fig. 5, a wider attenuation band
resulted for the proposed filter when compared
with CSRRs. This demonstrates the sharp and
strong suppression of the proposed bandstop filter.
Fig. 7 (a) Complementary spiral resonator (CSR) unit cell

with its relevant dimensions, and (b) the proposed slottedCSR unit cell. Note that gray area represents metallization.
Fig. 6 Attenuation coefficient for the proposed filter with SCSRRs and compared with CSRRs filter (without the slot).
B. Slotted-Spirals based filters
The geometrical dimensions of the CSRs are: side

length L = 6 mm, a = 0.2 mm, b = 0.4 mm, where a
dielectric substrate (r = 3.48, tan = 0.004) with
a thickness of 1.27 mm is used. The periodicity,
that is separation distance between adjacent CSR
elements, was chosen as 0.4 mm. Three CSR unit
cells are placed adjacent to each other and etched
out from metallic ground plane beneath a TL
segment of width 1.18 mm in order to match it to
50 , as shown in Fig. 8. Two ports were used to
numerically compute the transmission coefficient,
S21.
404
Fig. 9 Simulated transmission coefficient, S21, for the filter

with CSRs. Note that no slits are presented for the CSRs.
Fig. 8 3D view of the bandstop filter with complementary
spiral resonators etched out from metallic ground plane.
Note that TL segment is matched to 50.
Fig. 9 depicts the transmission coefficient of the

bandstop filter with three CSRs etched out from
the ground plane. A narrowband suppression for
the filter can be seen around 2.45 GHz, that makes
such filters suitable for wireless local area
networks to reject any kind of electromagnetic
noise around that particular frequency. It is
indeed expected to achieve such narrowband
behavior from such resonators, due to their
inherent high quality factors. Furthermore, Fig.9
shows the possibility of miniaturizing the filter by
just changing the host medium above the ground
plane with the etched CSRs, from r = 3.48 to r =
10.2, the suppression band is shifted to lower
frequencies, around 1.25 GHz in addition to
another suppression band around 3.2 GHz.
Fig. 10 shows snapshots for the computed surface
current distribution along the metallic ground
layer of this filter with CSRs. It is observed that
high current concentration is observed at the
resonant frequencies of the CSRs that lie within
the suppression band of the filter, at 1.2 GHz and
3.0 GHz as shown in Figs. 10(a) and (c). On the
other hand, low current circulates around the
resonators (see Fig. 10(b)) at the non-resonant
frequency, 2.5 GHz as expected. Note that these
snapshots correspond to the filter with a host
dielectric medium, r = 10.2, tan = 0.0023.
Fig. 10 Snapshots for simulated surface current distribution

along the vicinity of the metallic ground plane of the
bandstop filter with CSRs at various frequencies: (a) 1.2
GHz, (b) 2.5 GHz, and (c) 3.0 GHz. Note that no slits are
presented for the CSRs.
405
The effect of the bridged slot between resonant
CSRs is investigated next on the suppression band
of bandstop filters. Similar model (see Fig. 11) to
that presented in Fig. 8 is numerically studied and
simulated to extract the suppression band of the
filer. The transmission coefficient, S21, is
computed by considering two ports at the edges of
a TL segment of width 3.5 mm for proper match to
a 50- impedance. The width and length of the
slot, that is used to bridge two CSRs, is chosen as
Ws = 0.15 mm and Ls = 2 mm, while the length of
the two identical CSRs is taken as before. The
results are presented as shown in Fig. 12. As can
be seen, the effect of the slotted bridge has
resulted in a dramatic increase in the suppression
band of the filter, with almost 2 GHz band of
signal rejection when compared with the filter
without the slotted line (see Fig. 9). Furthermore,
the effect of the host dielectric material is noted to
further miniaturize the bandstop filter as can be
seen from Fig. 12.
Fig. 11 3D view of the bandstop filter with slottedcomplementary spiral resonators etched out from metallic
ground plane. Note that TL segment is matched to 50.
For this particular study, the effect of the dielectric

material is considered with only two host media.
Next, the current distribution for the filter with
slotted-CSRs is shown in Fig. 13 at the SCSRs
resonant frequency of 2.1 GH, when dielectric
medium (r = 3.48, tan = 0.002) is used.
Fig. 13 Snapshot for simulated surface current distribution

along the vicinity of the metallic ground plane of the
bandstop filter with Slotted-CSRs at resonant frequency of
2.1 GHz.
3 CONCLUSIONS
In this paper, bandstop filters based on developed
slotted complementary split ring resonators and
slotted complementary spiral resonators were
proposed and investigated. The filters were
numerically studied using full-wave time-domain
simulator of CST Microwave Studio. Parametric
studies based on varying the filters geometrical
parameters as well as the host dielectric medium
were thoroughly conducted and several remarks
based on the suppression band of the proposed
filter were highlighted in the context of the
discussed results.
4 ACKNOWLEDGMENT
The author would like to thank Sultan Qaboos
University for providing the resources to
accomplish this research work. The author also
acknowledges the support from CST Microwave
Studio for providing trials to do the full-wave
modeling of the filters.
Fig. 12 Simulated transmission coefficient, S21, for the filter

with slotted-CSRs (S-CSRs).
406
5 REFERENCES
[1] M. Mandal and P. Mondal, Design of sharp-rejection,
compact, wideband bandstop filters, IET Microw.
Antennas Propag., Vol. 2, No. 4, pp. 389-393, 2008.
[2] F. Yang, K. Ma, Y. Qian, and T. Itoh, A uniplanar
compact photonic-bandgap (UC-PBG) structure and its
applications for microwave circuit, IEEE Trans. Micro.
Theory Tech., Vol. 47, No. 8, pp. 1509-1514, 1999.
[3] C. Hang, W. Deal, T. Qian, and T. Itoh, High
efficiency transmitter front-ends integrated with planar
an PBG, Asia-Pacific Microwave Conf. Dig., pp. 888894, Dec. 2000.
[4] A. Safwat, F. Podevin, P. Ferrari, and A. Vilcot,
Tunable band-strop defected ground structure
resonator using reconfigurable dumbbell-shaped
coplanar waveguide, IEEE Trans. Microw. Theory
Tech., Vol. 54, No. 9, pp. 3559-3564, Sep. 2006.
[5] X. Chen, X. Shi, Y. Guo, and M. Xiao, A novel dual
band transmitter using microstrip defected ground
structure, Progress In Electromagnetics Research, Vol.
83, pp. 1-11, 2008.
[6] R. Marques, F. Martin, and Sorolla, M., Metamaterials
with negative parameters: theory, design and microwave
applications, John Wiley and Sons Inc., 2008.
[7] R. Marques, F. Mesa, J. Martel, and F. Medina,
Comparative analysis of edge- and broadside- coupled
split ring resonators for metamaterial design - theory
and experiments, IEEE Transactions on Antennas
Propagation, Vol. 51, Issue 10, pp. 2572-2581, Oct.
2003.
[8] J. Bonache, M. Gil, I. Gil, J. Garcia-Garcia, and F.
Martin, On the electrical characteristics of
complementary metamaterial resonators, IEEE
Microwave Wireless Components Letters, Vol. 16, Issue
10, pp. 543-545, Oct. 2006.
[9] J.B. Pendry, A.J. Holden, D.J. Robbins, and W.J.
Stewart, Magnetism from conductors and enhanced
nonlinear phenomena, IEEE Transactions on
Microwave Theory Techniques, Vol. 47, Issue 11, pp.
20752084, Nov 1999.
[10] D. R. Smith, W. Padilla D. Vier, S. Nemat-Nasser, S.
Schultz, Composite medium with simultaneously
negative permeability and permittivity, Phys. Rev.
Lett., 84, No. 18, pp. 4184, May 2000.
[11] V. Veselago, The electrodynamics of materials with
negative permittivity and negative permeability, Soviet
Physics USPEKI, 10, 509, 1968.
[12] F. Falcone, T. Lopetegi, J.D. Baena, R. Marques, F.
Martin, and M. Sorolla, Effective negative- stopband
microstrip lines based on complementary split ring
resonators, IEEE Microwave Wireless Components
Letters, Vol. 14, Issue 6, pp. 280-282, 2004.
[13] F. Falcone, T. Lopetegi, M. A. G. Laso, J. D. Baena, J.
Bonache, M. Beruete, R. Marques, F. Martin, and M.
Sorolla, Babinet principle applied to the design of
metasurfaces and metamaterials, Phys. Rev. Lett., Vol.
93, No. 19, pp. 197401-197404, Nov 2004.
[14] M.M. Bait-Suwailam, O.F. Siddiqui, and O.M. Ramahi,

Mutual coupling reduction between microstrip patch
antennas using slotted-complementary split-ring
resonators, IEEE Antenna Wireless Propagation
Letters, Vol. 9, pp. 876-878, 2010.
[15] M. M. Bait-Suwailam, Numerical study of bandstop
filters based on slotted-complementary split-ring
resonators (SCSRRs), in the second international
conference of technological advances in electrical,
electronics and computer engineering, pp. 34 37,
Kuala Lumpur, Malaysia, March 18-20.
[16] J. Garcia-Garcia, F. Martin, F. Falcone, J. Bonache,
J.D. Baena, I. Gil, E. Amat, T. Lopetegi, M.A.G. Laso,
J.A.M. Iturmendi, M. Sorolla, and R. Marques,
Microwave filters with improved stopband based on
sub-wavelength resonators, IEEE Trans. Microw.
Theory Tech., Vol. 53, No. 6, pp. 1997-2006, June 2005.
407

IJDIWC - Volume 4, Issue 3

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

IJDIWC - Volume 4, Issue 3

Uploaded by

Copyright:

Available Formats

CLOUD-LINKED AND CAMPUS-LINKED STUDENTS PERCEPTIONS Sivachandran Chandrasekaran,

AN APPLICATION OF DBR TO THE DEVELOPMENT OF A

AN ACTIVE MIDDLEWARE FOR SECURE AUTOMATIC S.

W.N.N.W. Marzudi, Z.Z. Abidin,

LITERATURE-BASED AND ARCHITECTURE-BASED MODERNIZATION

PERSPECTIVE FOR SUGGESTION ABOUT COMMUNICATION PLAN /

AN INSTRUCTIONAL DESIGN MODEL AND CRITERIA FOR Mohamed Elsayed

APPLICATIONS OF THE ROTATING CALIPERS TO GEOMETRIC

Tiago Cinto, Harlei M. A. Leite,

Cloud-Linked and Campus-Linked Students Perceptions of

issues. In project-oriented design based learning

Figure 1. Design based learning process

By engaging students in learning design, DBL

active learning process which makes students

The learning and assessment design, learning

Figure 2. The PODBL cycle

cloud and located learning activities. The students

cloud-linked and campus-linked students

Good experience, Time efficient

The work is divided, faster to

Figure 3: Modes of Design based learning preferred

The quantitative analysis of the survey results

Table 2 illustrates students perceptions on

No individual effort, loss of time

Hard to make decisions and

5.2 Students Perceptions on Design Based

When students are asked about the influence of

Table 3 shows students perception of design based

Figure 5: DBL to your career

It is clearly shown in figure 5, campus-linked

Practical Learning, Demos in

Table 4 indicates cloud-linked/campus-linked

Laal, M., and Laal, M.: Collaborative learning: what is

23. Doppelt, Y., Schunn, C.D.: Identifying students

An Application of DBR to the Development of

line with Japanese government plans to employ

DBR, with its flexible and practical characteristics

problems in the placement test the university had

and using the data for educational purposes after

minimum standard. By using similar entrance

P(Dj|H1)P(H1) > P(Dj|H2)P(H2) > P(Dj|H3)P(H3)

P(Dj|H1)P(H1) < P(Dj|H2)P(H2) > P(Dj|H3)P(H3)

P(Dj|H1)P(H1) < P(Dj|H2)P(H2) < P(Dj|H3)P(H3)

In the equations, P(H1), P(H2), and P(H3) are prior

The e-learning system we developed also used the

SET001 to SET012 (99.1%66.4%)

During their two years of compulsory English

Figure 2. Example of Basic test items

2.3 Phase 3: Testing and refinement of solutions

Figure 3. Result of the test.

Figure 4. Person and item map

In Phase 4, the placement test (ver. 1.0; N = 653,

Figure 5. Result of the placement test ver. 1.0

budgetary and manpower savings will thus be

Figure 6. Result of the placement test and the

Figure 6 shows the result of 31 shared test items in

In regard to the third point, we managed to

Certain limitations to our efforts did, however,

are confident the integrated framework we

Japanese Ministry of Education, Culture, Sports,

the Research Agenda. Educational Technology &

An Active Middleware for Secure Automatic Reconfiguration of Applications for