INTERNATIONAL CONFERENCE ON CURRENT INNOVATIONS IN ENGINEERING AND TECHNOLOGY

ISBN: 378 - 26 - 138420 - 5

A Secured Based Information Sharing Scheme via

Smartphones in DTN Routings
T. Mohan Krishna1 V.Sucharita2
2

PG Scholar1, M.Tech. Associate Professor

Department of Computer Science and Engineering,
mohan.gkce@gmail.com
Audisankara College of Engineering & Technology, Gudur.
A.P - India
Abstract :

With the growing range of smart phone users, peer-to-peer content sharing is anticipated to occur a lot of usually. Thus, new
content sharing mechanisms ought to be developed as ancient information delivery schemes aren't economical for content
sharing attributable to the isolated property between sensible phones. To achieve data delivery in such complex environments,
researchers have anticipated the use of epidemic routing or store-carry-forward protocols, in this a node stores a message may
be a note and carries it until a forwarding chance arises through an encounter with other node. Earlier studies in this field
focused on whether two nodes would come across each other and the place and time of encounter.During this paper, we tend to
propose discover-predict-deliver as associate economical content sharing theme for delay-tolerant Smartphone networks. In our
projected theme,. Specifically, our approach employs a quality learning formula to spot places inside and outdoors. A hidden
markov model and viterbi algorithm is employed to predict associate individuals future quality info. analysis supported real
traces indicates that with the projected approach, eighty seven p.c of contents is properly discovered and delivered among a pair
of hours once the content is out there solely in thirty p.c of nodes within the network .In order to decrease energy consumption
we are using asymmetric multi core processors and efficient sensor scheduling is needed for that purpose we are using POMPDs
in sensor scheduling.
Keywords:- Tolerant Network, ad hoc networks, Store and forward networks, peer-to-peer network
I.INTRODUCTION
Now a days Number of advanced smart phone
clients has rapidly expanded in the recent years. users can
make different sorts of substance effectively utilizing easy to
use interfaces accessible within the advanced smart phones
phones. However substance offering among shrewd
telephones is dreary as it obliges a few activities, for example,
transferring to concentrated servers, seeking and downloading
substance. One simple route is to depend on adhoc technique
for distributed substance imparting. Shockingly with the
current adhoc steering conventions, substance are most
certainly not conveyed if a system parcel exists between the
associates at the point when substance are imparted. Thus
Delay Tolerant System (DTN) steering conventions
accomplish better execution than customary adhoc steering
conventions. These conventions don't oblige an incorporated
server. Thus the substance are put away on advanced mobile
phones itself. Smartphone's consists of many network
interfaces like Bluetooth and Wi-Fi so ad hoc networks can be
easily constructed with them. The Connectivity among
Smartphone's is likely to be alternating because of movement
patterns of carriers and the signal transmission phenomena.
A wide variety of Store-carry-forward protocols have been
anticipated by researchers. Routing in delay-tolerant
networking concerns itself with the ability to route, data from

a source to a destination, which is a essential ability of all

communication networks must have. In these exigent
environments, mostly used or familiar ad hoc routing
protocols fail to establish routes. This is due to these
protocols first try to establish a complete route and then, once
the route has been established forwards the actual data. Still,
when immediate end-to-end paths are complicated or
unfeasible to institute, routing protocols should take to a
"store and then forward" method or approach, where data or a
message is moved and stored incrementally all over the
network in hops that it will finally arrive at its destination. A
general technique used to maximize the likelihood of a
message being effectively transferred is to duplicate many
copies of the message in hops that one will be successful in
reaching its destination. DTN workplace of forty buses and
simulations supported real traces. we tend to concentrate on
store-carry-forward networking situations, during which the
nodes communicate victimization DTN bundle design. Some
smart phones within the network store content that they're
willing to share with others. All smart phone users square
measure willing to get together and provide a restricted
quantity of their resources, like information measure, storage,
and process power, to help others. Our goal is to permit users
to issue queries for content hold on on alternative smart
phones anyplace within the network and to assess the
probabilities of getting the knowledge required. we tend to

INTERNATIONAL ASSOCIATION OF ENGINEERING & TECHNOLOGY FOR SKILL DEVELOPMENT

107

www.iaetsd.in

INTERNATIONAL CONFERENCE ON CURRENT INNOVATIONS IN ENGINEERING AND TECHNOLOGY

assume that smart phones will perform searches on their

native storage, and that we realize the relevant results for a
given question to facilitate looking.
Delay Tolerant Network (DTN) routing
protocols attain enhanced performance than usual ad hoc
routing protocols. Over the proposed DTN routing protocols,
Epidemic routing is a essential DTN routing solution. In
Epidemic routing by vahdat et al[2], messages are forwarded
to each encountered node that does not have a replica of the
same message. This solution exhibits the finest performance
in terms of delivery pace and latency, but it involves abundant
resources, such as storage, bandwidth, and energy.
This paper spotlight mainly on efficiency of content
discovery and its delivery to the targeted destination. Here we
suggest recommendation based discover-predict-deliver
(DPD) as efficient and effective content sharing scheme for
smart phone based DTNs. DPD suppose that smart phones
can hook up when they are in close proximity that is where the
Smartphone users reside for a longer period. Earlier studies
have shown that Smartphone users stay indoors for a longer
period where GPS cannot be accessed. The objective of our
work is to discover solutions to the problems in content
sharing and to minimize the energy consumption using sensor
scheduling scheme and finally improve the performance and
want to reduce energy consumption by using asymmetric
multicore processors and efficient sensor scheduling my using
POMPDs in scheduling scheme.

ISBN: 378 - 26 - 138420 - 5

rule downside and conjointly gift many extensions to permit

OPF to use solely partial routing data and work with different
probabilistic forwarding schemes like ticket-based forwarding.
Implement OPF and several and a number of different and
several other other protocols and perform trace-driven
simulations. Simulation results show that the delivery rate of
OPF is barely five-hitter below epidemic, and two hundredth
bigger than the progressive delegation forwarding whereas
generating five-hitter a lot of copies and five-hitter longer
delay.
B.DTN Routing as a Resource Allocation Problem:
Many DTN routing protocols use a range of mechanisms, as
well as discovering the meeting possibilities among nodes,
packet replication, and network cryptography. The first focus
of those mechanisms is to extend the chance of finding a path
with restricted info, thus these approaches have solely
associate incidental result on such routing metrics as most or
average delivery latency. during this paper we have a
tendency to gift fast associate intentional DTN routing
protocol which will optimize a particular routing metric like
worst-case delivery latency or the fraction of packets that area
unit delivered inside a point. The key insight is to treat DTN
routing as a resource allocation drawback that interprets the
routing metric into per-packet utilities that confirm however
packets ought to be replicated within the system.
C.Resource Constraints:
RAPID (resource allocation protocol for intentional DTN)
routing a protocol designed to expressly optimize AN
administrator-specified routing metric. speedy routes a
packet by opportunistically replicating it till a replica reaches
the destination. Speedy interprets the routing metric to perpacket utilities that verify at each transfer chance if the utility
of replicating a packet justifies the resources used. Speedy
loosely tracks network resources through an impact plane to
assimilate a neighbor hood read of worldwide network state.
To the present finish speedy uses AN in-band management
channel to exchange network state data among nodes
employing a fraction of the obtainable information measure.
D. Epidemic routing:

Fig1: Finding meaning ful places and their simulation area

II.LITERATURE SURVEY
A. Optimal Probabilistic Forwarding Protocol in DTN:
To provide Associate in Nursing optimum forwarding
protocol that maximizes the expected delivery rate whereas
satisfying a definite constant on the amount of forwardings
per message The optimum probabilistic forwarding (OPF)
protocol, we have a tendency to use Associate in Nursing
optimum probabilistic forwarding metric derived by modeling
every forwarding as Associate in Nursing optimum stopping

Epidemic routing may be a easy answer for DTNs, during

which messages square measure forwarded to each
encountered node. Thus, Epidemic routing achieves the best
attainable delivery rate and lowest attainable latency, however
it needs vast information measure and storage resources.
Investigated a settled wave model for the progress of
Epidemic routing. Many approaches are projected to cut back
the overhead and to boost the performance of Epidemic
routing examined variety of various ways to suppress
redundant transmissions. projected conversation strategies,
during which a node chooses a random range between zero
and one, and therefore the message is forwarded to a different

INTERNATIONAL ASSOCIATION OF ENGINEERING & TECHNOLOGY FOR SKILL DEVELOPMENT

108

www.iaetsd.in

INTERNATIONAL CONFERENCE ON CURRENT INNOVATIONS IN ENGINEERING AND TECHNOLOGY

node if the chosen range is higher than a user-predefined

chance. These works belong to resource- aware routing
protocols. different protocols square measure classified into
two groups. Opportunity-based schemes and prediction-based
schemes.

III.RELATED WORK
A delay tolerant network (DTN) is a mobile
network where a existing source-destination path may not
exist amid a pair of nodes and messages are forwarded in a
store-carry-forward routing hypothesis [6]. Vahdat et al [2]
anticipated Epidemic routing as fundamental DTN routing
protocol in which a node forwards a message to each and
every encountered node that does not have a replica of the
message. The solution shows the finest performance terms of
delivery pace and latency but wastes bulk of bandwidth.
A substitute solution was resource based [3], [4],where
systems utilize data mules as message carriers that straightly
delivers the message to the destination. Next, opportunitybased routing protocols make use of history of encounters to
convey a message to the destination [5], [6], [7]. Prediction
based schemes [8], [9], make use of complicated utility
functions to decide whether the forward a message to the
node.

ISBN: 378 - 26 - 138420 - 5

will be communicating using the DTN bundle protocol. A few

devices in the network store content which they are ready to
share with others. All nodes are willing to assist and provide a
restricted amount of their local system resources (bandwidth,
storage, and dispensation power) to aid other nodes. Our
objective is to permit users to issue queries for content that is
stored on the other nodes everywhere in the network and
consider the possibility of such a node to acquire the required
information. To ease searching, we suppose that nodes are
capable to carry out searches on their local storage and
uncover the appropriate results for a given query. The content
sharing process is characterized into two phases:
content discovery phase and the content delivery phase.
In the content discovery phase, the user inputs or enters in a
content sharing application requests for the content. The
application initially searches the content it its own or
individual database and if not found, the application then
creates a query that is forwarded based on the users request.
When the content is found, the content delivery phase is
initiated or commenced, and the content is forwarded to the
query originator.

Yuan et al precisely predicted encounter opportunities by

means of time of encounters. Pitkanen et al anticipated stateof-the-art content sharing scheme in DTNs. They mainly
focused on restraining search query propagation and
anticipated several query processing methods.
Chang et al anticipated a process for searching for a node or
an object in a bulky network and restraining search query
propagation. Here a class or a set of controlled flooding
search strategies where query/search packets are broadcasted
and propagated in the network in anticipation of a preset TTL
(time-to-live) value carried in the packet terminates. The
objective of our work is to discover the content sharing
problem in Smartphone based DTNs involves minimizing
energy consumption using sensor scheduling schemes.
In General, these works are outlined focused around contact
histories. when two or more smart phone clients are in the
same spot, their gadgets don't generally make correspondence
or recognize contact opportunities. Subsequently, the contact
history gives less exact data on future contact opportunities
than portability data. This result might be seen in the dissects
talked about
Content sharing in DTN'S involves the following problems:
A. Content sharing
In this segment we examine the problem of content
sharing in delay tolerant networks and depict substitute
solutions. As specified in the introduction, we spotlight on
mobile opportunistic networking scenarios where the nodes

Figure: Processing of incoming query.

B. Content discovery
In content discovery, mainly systems spotlight on how to
formulate queries, that depends on assumptions about the
format or layout of the content to be discovered. A common
protocol should sustain various types of queries and content,
but we abstract or summarize from the actual similar or
matching process in order to spotlight on discovering content
in the network. The easiest strategy to discover and deliver
the contents is Epidemic routing [2]. But, due to resource
limitations, Epidemic routing is often extravagant, so we have
to consider methods that limits the system resources used up
on both content discovery and delivery. Preferably, a query
should only be forwarded to neighbours that hold on the
matching contents or those are on the path to other nodes
having matching content . Different nodes should return no
overlapping responses to the requester. As total knowledge or
active coordination is not an alternative in our scenario, each

INTERNATIONAL ASSOCIATION OF ENGINEERING & TECHNOLOGY FOR SKILL DEVELOPMENT

109

www.iaetsd.in

INTERNATIONAL CONFERENCE ON CURRENT INNOVATIONS IN ENGINEERING AND TECHNOLOGY

node can only make autonomous forwarding decisions. These

autonomous forwarding decisions should attain a fine trade off
amid discovery efficiency and necessary resources. Analogous
limitations pertain to content delivery. A few methods
anticipated by Pitkanen et al. may be used for restraining the
distribution of queries. Additionally, we study two substitutes
for restraining the distribution of queries: a split query lifetime
limit and a query distance limit. We employ the controlled
replication-based [9] routing scheme that performs a singlecopy scheme. This single-copy scheme turn both query
lifetime and distance limits into random walk, and the scheme
is not effectual when content-carrier nodes (i.e., destinations)
are not well-known. By distinguishing, the controlled
replication-based scheme dispenses a set of message replicas
and evade the excessive spread of messages.
C. Content delivery
When the query matching content is discovered, the
content carrying node should transmit only a subset of results.
This constraint is needed to limit the amount of resources
utilized both locally and globally for sending and storing the
responses, and to eliminate potential copies . The query
originator sets a limit for both the number of replications or
duplicates and the amount of content that should be produced.
When nodes require to forward a query message, the limits
incorporated in the query message are used to make the
forwarding decision. If the amount of the content go beyond
the response limit, the node wants to select which ones to
forward.
D. Mobility Prediction
Numerous studies have largely specified another
problem of content sharing: mobility learning and prediction.
Beacon Print discover meaningful places by constantly
determining constant scans for a time period. Place Sense
senses the arrival and exit from a place by utilizing invasive
RF-beacons. The system uses a radio beacons retort rates to
attain vigorous beacon conclusion. EnTracked is a position
tracking system for GPS-enabled devices. The system is
configurable to recognize different tradeoffs amid energy
consumption and heftiness. Mobility prediction has been
extensively studied in and out of the delay-tolerant networking
area. Markov-based schemes, make the problem as a Hidden
Markov or semi-Markov model and probabilistic prediction of
human mobility. In contrast, neural network based schemes
try to match the observed user behaviour with earlier
observed behaviour and estimate the prospect based on the
experimental patterns.Markov based schemes are suitable for
resource- restricted devices, like Smartphones, owing to their
low computation overhead and reserved storage requirements.
In our work, we have to develop a mobility learning and
prediction method. This method has been built to offer coarsegrained mobility information with a less computation
overhead. When the difficulty of mobility learning and

ISBN: 378 - 26 - 138420 - 5

prediction scheme can be mistreated, the schemes specified in

can be worn to offer fine-grained mobility information.

Problem in previous works

In the previous works, the energy consumption is more so
the battery lifetime will be reduced. By using asymmetric
multicore processors and efficient sensor scheduling
mechanisms energy consumption can be reduced and can
increase the lifespan of the batteries.
IV. PROPOSED WORK
In order to content sharing we can use the DPD Technique
and with that Here we are using asymmetric multicore
processors for performance expansion in sensing. Below fig
shows the daily energy utilization outline, which is calculated
using the composed stationary and movement time from
seven different users in four weeks . The scrutiny does not
comprise the energy utilization of content swap as it mainly
rely on the volume and communication pace of the nodes. The
typical energy consumptions of GPS, Wi-Fi, and the
accelerometer varies. The accelerometer do have
the
maximum energy consumption as it is used endlessly over 24
hours. Wi-Fi energy utilization is observed owed to the
scanning of neighbour APs for place recognition. GPS has a
huge discrepancy in energy consumption as this may not be
available in all places.
In this paper we examine the issue of following an item
moving random through a system of remote sensors. Our
destination is to devise techniques for planning the sensors to
advance the trade off between following execution and energy
consumption. We give the scheduling issue a role as a
Partially observable Markov Decision Process (POMDP),
where the control activities compare to the set of sensors to
initiate at each one time step. Utilizing a bottom-up approach
methodology, we consider different sensing, movement what's
more cost models with expanding levels of trouble
In order to minimize energy consumption or utilization
we use sensor scheduling schemes or mechanisms[10]. Sensor
systems have an wide-ranging diversity of prospective,
functional and important applications. In any case, there are
questions that have to be inclined for prolific procedure of
sensor system frameworks in right applications. Energy
sparing is one fundamental issue for sensor systems as most of
the sensors are furnished with no rechargeable batteries that
have constrained lifetime. To enhance the lifetime of a sensor
set up, one vital methodology is to attentively schedule
sensors' work sleep cycles (or obligation cycles). In addition,
in cluster based systems, grouping heads are usually selected
in a way that minimizes or reduces the aggregate energy
utilization and they may axle among the sensors to fine-tune
energy utilization. As a rule, these energy productive
scheduling components or mechanisms (furthermore called
topology arrangement components) required to accomplish

INTERNATIONAL ASSOCIATION OF ENGINEERING & TECHNOLOGY FOR SKILL DEVELOPMENT

110

www.iaetsd.in

INTERNATIONAL CONFERENCE ON CURRENT INNOVATIONS IN ENGINEERING AND TECHNOLOGY

certain application requirements while sparing energy. In

sensor arranges that have various outline requirements than
those in conventional remote systems. Distinctive instruments
may make characteristic suspicions about their sensors
together with identification model, sense zone, transmission
scope, dissatisfaction or disappointment model, time
management, furthermore the capability to get area and
parting data.

ISBN: 378 - 26 - 138420 - 5

discovery rate. With an increasing query lifetime, both DPD

and Epidemic show a high discovery ratio because with a
longer duration, each query is forwarded to more nodes.

Fig 4: Discovery Efficiency

The influence of the query lifetime on hop-based discovery
methods is not significant. These observations are derived
from the limitation on the number of forwards.
C. Prediction Accuracy;
Figure: Mean energy consumption in a day
V.PERFORMANCE EVALUATION
A. Learning Accuracy
Learning accuracy shows how efficiently and correctly the
places were identified. The accuracy of place learning
influences the estimation of encounter opportunity between
two nodes. For example, If two different places are identified
as identical ones, we may incorrectly estimate that two nodes
will encounter each other when they visit two different
places.Also, the correct computation of and depends on the
geographical location information of nodes.

Mobility prediction is a key factor in the estimation of utility

function. Here, we evaluate our prediction method according
to trajectory deviation, prediction duration, and learning
period, as illustrated. Trajectory deviation indicates the
irregularity of a users mobility. For this evaluation, we
modify the existing mobility information with noise data.
Thus, 10, 20, and 30 percent of the meaningful places are
randomly chosen locations for trajectory deviations of 0.1,
0.2, and 0.3, respectively. As the trajectory deviation
increases, the prediction accuracy decreases. Prediction
accuracy is computed as the ratio of correctly predicted
locations to the total predicted locations.

Fig3: Learning Accuracy

Fig 5: Prediction Accuracy

B. Discovery Efficiency

D. Sharing Cost and Latency:

The discovery ratio is the ratio of discovered contents to the

generated queries within a given duration. DPDs discovery
performance is subjective to the two forwarding. In Epidemic,
queries are forwarded to every node. In hops-10 and hops-5, a
query message is forwarded until its hop count reaches 10 and
5, respectively. When a query matching content is available
only on a few nodes, the discovery methods show a low

Finally, we evaluate the protocols in terms of latency and cost,

as shown in Fig. 3f and Fig. 4. E-E uses Epidemic routing in
both the discovery and delivery phases. E-S&W uses
Epidemic routing for content discovery and Spray and Wait
for content delivery. The sharing latency is the sum of the
discovery latency and the delivery latency. E-E shows the

INTERNATIONAL ASSOCIATION OF ENGINEERING & TECHNOLOGY FOR SKILL DEVELOPMENT

111

www.iaetsd.in

INTERNATIONAL CONFERENCE ON CURRENT INNOVATIONS IN ENGINEERING AND TECHNOLOGY

lowest latency, and both DPD and E-S&W show the highest
latency. DPD exhibits such results due to the high latency of
the discovery phase. However, the delivery latency of DPD is
much smaller than that of E-S&W and is close to that of E-E.
E-E shows the highest overhead. The latency and overhead are
tradeoffs. In summary, DPD achieves good efficiency in the
delivery phase, whereas the efficiency of the discovery phase
can be improved. Content header caching on all nodes may be
a good solution, and this issue will be addressed in future
works.

ISBN: 378 - 26 - 138420 - 5

REFERENCE
[1]

T3I Group LLC, http://www.telecomweb.com, 2010.

[2]

A. Vahdat and D. Becker, Epidemic Routing for Partially Connected

Ad Hoc Networks, technical report, Dept. of Computer Science, Duke
Univ., Sept. 2000.

[3]

A. Balasubramanian, B.N. Levine, and A. Venkataramani, DTN

Routing as a Resource Allocation Problem, Proc. ACM SIGCOMM,
pp. 373-384, 2007.

[4]

R.C. Shah, S. Roy, S. Jain, and W. Brunette, Data Mules: Modelling a

Three-Tier Architecture for Sparse Sensor Networks, Elsevier Ad Hoc
Networks J., vol. 1, pp. 215-233, Sept. 2003.

[5]

A. Lindgren, A. Doria, and O. Schelen, Probabilistic Routing in

Intermittently Connected Networks, SIGMOBILE Mobile Computer
Comm. Rev., vol. 7, no. 3, pp. 19-20, 2003.

[6]

C. Liu and J. Wu, An Optimal Probabilistic Forwarding Protocol in

Delay Tolerant Networks, Proc. ACM MobiHoc, pp. 14, 2009.

[7]

J. Wu, M. Lu, and F. Li, Utility-Based Opportunistic Routing in MultiHop Wireless Networks, Proc. 28th Intl Conf. Distributed Computing
Systems (ICDCS 08), pp. 470-477, 2008.

[8] T. Spyropoulos, K. Psounis, and C.S. Raghavendra, Spray and Wait: An

Efficient Routing Scheme for Intermittently Connected Mobile
Networks, Proc. ACM SIGCOMM Workshop Delay-Tolerant
Networking (WDTN 05), pp. 252-259, 2005.

Fig 4: Sharing Cost and Latency

CONCLUSION

[9]

In this paper, we tend to planned associate degree economical

content sharing mechanism in Smartphone-based DTNs. We
tend to tried to utilize the benefits of todays Smartphones
(i.e., handiness of varied localization and communication
technologies) and suitably designed the protocol. In planning
a content sharing algorithmic rule, we tend to centered on 2
points: 1) individuals move around meaningful places, and 2)
the quality of individuals is certain. supported this proposition,
we tend to developed a quality learning and prediction
algorithmic rule to reckon the utility operate. We tend to
learned that contents so have geographical and temporal
validity, and that we planned a theme by considering these
characteristics of content. for instance, distributing queries for
content in a vicinity twenty miles from the placement of the
content searcher has solely a zero. Three percentage
probability to get the content whereas generating twenty
percentage additional transmission price. the time limitation
on question distribution reduces transmission price. most vital,
the planned protocol properly discovers and delivers eighty
seven percentage of contents among two hours once the
contents square measure on the market solely in thirty
percentage of nodes within the network.energy consumption
are reduced by using sensor scheduling and further works
have to done on user privacy.

T. Spyropoulos, K. Psounis, and C. Raghavendra, Efficient Routing in

Intermittently Connected Mobile Networks: The Single-Copy Case,
IEEE/ACM Trans. Networking, vol. 16, no. 1, pp. 63-76, Feb. 2008.

[10] T. Spyropoulos, K. Psounis, and C.S. Raghavendra, Efficient Routing in

Intermittently Connected Mobile Networks: The Multiple-Copy Case,
IEEE/ACM Trans. Networking, vol. 16, pp. 77-90, Feb. 2008.
[11] I. Cardei, C. Liu, J. Wu, and Q. Yuan, DTN Routing with Probabilistic
Trajectory Prediction, Proc. Third Intl Conf. Wireless Algorithms,
Systems, and Applications (WASA 08,), pp. 40-51, 2008.
[12] Q. Yuan, I. Cardei, and J. Wu, Predict and Relay: An Efficient
Routing in Disruption-Tolerant Networks, Proc. 10th ACM MobiHoc,
pp. 95-104, 2009.
[13]

E.M. Daly and M. Haahr, Social Network Analysis for Routing in

Disconnected Delay-tolerant MANETs, Proc. Eighth ACM MobiHoc,
pp. 32-40, 2007.

[14]

N.B. Chang and M. Liu, Controlled Flooding Search in a Large

Network, IEEE/ACM Trans. Networking, vol. 15, no. 2, pp. 436- 449,
Apr. 2007.

[15]

C. Avin and C. Brito, Efficient and Robust Query Processing in

Dynamic Environments Using Random Walk Techniques, Proc. Third
Intl Symp. Information Processing in Sensor Networks (IPSN 04), pp.
277-286, 2004.

[16]

M. Pitkanen, T. Karkkainen, J. Greifenberg, and J. Ott, Searching for

Content in Mobile DTNs, Proc. IEEE Intl Conf. Pervasive Computing
and Comm. (PERCOM 09), pp. 1-10, 2009.

INTERNATIONAL ASSOCIATION OF ENGINEERING & TECHNOLOGY FOR SKILL DEVELOPMENT

112

www.iaetsd.in