1

Performance evaluation of a Volcano Monitoring

System Using Wireless Sensor Networks
Roman Lara-Cueva, Member, IEEE, Antonio Caamano, Member, IEEE, Marco Zennaro, Member, IEEE
and Jose Luis Rojo-Alvarez, Senior Member, IEEE

AbstractWireless Sensor Network have become critical in

the evolution of Telecommunications. Our interest is related to
monitor an Active Volcano using wireless sensor networks where
the requirement of real-time is mandatory, due to the necessity
of access immediately of the signals derived from a natural
disaster in order to determine emergency early warnings. Our
aim was to determine the number of sensors will be deployed
in a Volcano Monitoring System based on simulation results and
corroborated with an in-situ testbed. We used ns-2 as simulation
tool, where two scenarios were evaluated. This study determined
the optimal scenario in volcano monitoring is Randomly with
maximum eighteen nodes to satisfy metrics as throughput, time
delay and packet loss. We deployed sixteen sensors in a strategy
area at Cotopaxi Volcano, the information was obtained during
three days of continuous monitoring. This information was sent
to a surveillance laboratory located 45 km away from the station
placed at the volcano, a WiFi-based long distance technology was
used to this purpose. Volcanic information was processed in time,
frequency and scale domain, a spectral pattern of seismic events
determined four kinds of events, corresponding to a long period
events, volcano tectonic earthquakes, volcanic tremor, and hybrid
events.
Index TermsWSN, 802.15.4, throughput, delay, packet loss,
monitoring system, volcano applications.

I. I NTRODUCTION TO WSN
Wireless Sensor Network (WSN) have become critical in
the evolution of Telecommunications, its constant evolution
permits the possibility to implement devices with low cost and
energy autonomy, without periodic maintenance, to be capable
of obtaining environmental information, and for these reasons
Cyber-Physical Systems (CPS), Internet of Things (IoT), and
Smart Cities are new research topics based on technologies
like WSN, all of them are based in a similar infrastructure of
every heterogeneous networks where data must be transmitted,
processed, and finally enable people through any application
to monitor or to control objects [1][6].
Ecuador has a special interest to use WSN in volcano
monitoring applications, as it is located in the Pacific Ring
of Fire a place with high seismic activity. WSN systems are
Manuscript received July 5, 2014
R.A. Lara is with Wireless Communication Research Group (WiCOM)
and Ad Hoc Networks Research Center (CIRAD) of the Department of Electrical and Electronic Engineering, Universidad de las
Fuerzas Armadas ESPE, Sangolqu-Ecuador, 171-5-231B e-mail: (see
http://wicom.espe.edu.ec/contactos.html).
A. Caamano and J.L. Rojo are with Information and Communication
Technology Department, Rey Juan Carlos University, Camino del Molino s/n
28943, Fuenlabrada-Madrid, Espana
M. Zennaro are with ICT for Development Laboratory , The Abdus Salam
International Centre for Theorical Physics, Trieste-Italy

much cheaper and reliable than bulky and energy-hungry traditional systems. Currently, South America lacks of permanent
monitoring systems deployed in active volcanoes, but some
WSN-based systems have been installed just in a couple of
weeks. Volcano monitoring using WSN still requires further
research in order to present information in real-time and to
launch an early emergency warning. The main constraint to
be considered as real-time systems is the time delay caused
by processing data, there are some solutions at Network and
MAC layer referred to data acquisition in-situ, data gathering
and data dissemination which significantly reduces time delay,
but it is impossible to give an early warning with this kind of
systems, because data are processed off-line in a far distanced
surveillance laboratory. The time delay related to digital signal
processing and the signal propagation must be solved within
appropriate processing and telecommunication techniques [7]
[9].
Our aim was to determine a number of sensors that maximize the network capacity, we considered as main metrics
throughput, time delay and packet loss in the WSN, and
then corroborated by an in-situ testbed deployed at Cotopaxi
volcano [10].
The rest of the paper is organized as follows. Section
2 summarizes previous research on the subject. Section 3
describes the performance evaluation of the WSN in detail.
Sections 4 and 5 describe the simulation results obtained
and the experimental study performed, respectively. Finally,
Section 6 presents our conclusions and future work [11].
II. R ELATED W ORK
Our interest consist in determining the network behavior,
for that reason we must study the performance network, this
can be evaluated by some Quality of Service (QoS) metrics
such as: availability, reliability, response time, time delay,
throughput, bandwidth capacity, and packet loss ratio. In order
to offer real-time in WSN with guaranteed QoS metrics, the
network must be analyzed in a different way than traditional
real-time systems, as far as WSN requires to need severe
challenges due to its wireless nature, distributed architecture
and dynamic network topology. The state of the art of realtime solutions currently developed have been presented with
emphasis at level of MAC, routing, data processing, and cross
layer, this denote a direct relationship between real-time and
QoS metrics, as well as new general concepts related to realtime WSN systems [12] [13]. Real-Time (RT) WSN could be
defined as a WSN capable of ensuring a Maximum Sustained

300
250
200
2^(n+1) (%)

Traffic Rate (Throughput), a Minimum Latency, and Packet

Loss as main QoS metrics. An ideal development process
should start from the theoretical analysis of the protocol to
provide bounds and information about its performance [14],
[15], it must be verified and refined by simulations [16]
[21], and finally confirmed in a testbed [22], [23]. We found
several works which presented a mixed analysis, since in real
scenarios it is possible to obtain measures of main metrics as:
RSSI, PER, and EED through tools developed by manufactures
[24][26].

150
100
50
0
0.12
0.1

80

0.08

60

0.06
40

0.04
20

0.02
Delay (s)

III. P ERFORMANCE E VALUATION OF THE MAC PROTOCOL

A. Real-Time WSN for a Volcano Monitoring System Requirements
For our application we have to consider the environment
presented by a Volcano a wild terrain and a lack of energy
to implement a WSN. A mesh topology would present the best
way to communicate sensors in this kind of scenario, specially
because we have to define the position of the nodes according
to the requirements that an in situ visit could give us according
to the variables to monitor.
It is developed a research about topologies for ad hoc
networks and it is presented the concept of tessellation, in
where, this model could be applied in WSN to obtain regular
network formed by geometrical figures [27], that can be
specified using the notation of Schlafli [28].

Fig. 1.

Number of nodes (n)

XXXXXX

groups: general parameters, power parameters, and node parameters, the rest of the parameters to simulate the network
model are detailed in the Tables I, II and III respectively.
TABLE I
G ENERAL PARAMETERS
Parameter
Radio Propagation Model
Routing Protocol
Raw Bit Rate (kbps)
Antenna Type
Simulation Time (s)
Simulation Time (s)

Value
Two-Ray Ground
AODV
250
Directional
260
220

B. Simulation Environment
There is a wide range of simulators that can be used to test
WSN in order to obtain several results to be analyzed based
on [29] we have chosen ns-2 as simulation tool.
In order to obtain the performance of the network through
simulation we have chosen two scenarios: tessellation and
randomly, for the first one we have chosen a triangular
tessellation pattern network {3, 6}, in the second one we
defined a random position of the sensor nodes placed on the
plane at a distance of 30 meters each (typical mean value for
connection in practice).
The number of nodes (n) in the triangular tessellation could
be obtained as function of the number of layers C, this is,
n = 1 + 3C(C + 1).

(1)

In both scenarios we started with 6 nodes growing until

60 nodes in 6 nodes step each, we defined one coordinator,
all transmissions nodes were directed to the coordinator. We
assumed an event occurred with a duration equal to 220
seconds in an approximated area of 300 300 m.
In our simulation process it was necessary to determine
a number of replications to reduce the mean square error,
consequently, we defined for tessellation scenario to run 6
replications, and for randomly scenario the number of replications depends of the number of the nodes in the scenario,
in other words, we run n replications in each scenario.
The main simulation parameters have been defined among
others: time simulation, topology, routing protocol, transmission rate, etc., we could cluster all of them in three main

TABLE II
P OWER PARAMETERS
Parameter
Transmission Power (dBm)
Sensitivity (dBm)
Transmission antenna gain Gt (dB)
Reception antenna gain Gr (dB)
Trajectory loss (dB)

Value
0 (1mW)
-94
1.0
1.0
1.0

TABLE III
N ODES PARAMETERS
Parameter
Traffic type
Traffic direction
Package size
Number of Coordinators
Distance between nodes
Number of nodes
Beacon mode

Value
FTP
all to Coordinator
55 bytes
1 coordinator
30 m
1 to 45 nodes
Enabled
Beacon Order:3
Superframe Order:3

The scenarios of both topologies of the networks defined for

our simulations are shown in Fig. 2, to obtain the results of
network performance and verify the operation with an optimal
number of sensors.
C. Performance Metrics
We selected three main metrics required for a real-time
monitoring: normalized throughput (), end-to-end delay

Tessellation topology

Normalized Throughput vs n
0.7

150
Coordinator
FFD Nodes

0.6

100

0.5

50

Position y (m)

0.4

0
0.3

50

0.2

0.1

100

150
150

100

50

0
Position x (m)

50

100

12

18

24

30

150

36
42
nodes (n)

48

54

60

66

48

54

60

66

Randomly topology
150

a) Tessellation Scenario

Coordinator
FFD Nodes

Normalized Throughput vs n
0.7

100

0.6

Position y (m)

50
0.5
0

0.4

50
0.3

100

150
150

0.2

0.1
100

50

0
Position x (m)

50

100

150
0

Fig. 2.

12

18

24

30

36
42
nodes (n)

Tessellation and Randomly Scenarios evaluated.

b) Randomly Scenario

(EED), and packet loss (PL). As mentioned in the previous

works there are other metrics that can be considered but we
could determine that all of them have a direct relation to our
main metrics, for example: duty cycle, energy consumption,
average jitter, load factor, and traffic type.
In several meetings maintained with experts in volcanology
from Instituto Geofsico de la Escuela Politecnica Nacional
(IGEPN), we determined that the system must be able to work
in a permanent way; to monitor several variables in a volcano it
is not effective to have a WSN in a saving power mode, for that
reason we did not consider power consumption metrics. For
those reasons, with respect to our metrics we need a maximum
, PL must be less than 20%, a minimum EED, and at least
5 stations must be necessaries. For we used the information
that permitted us to define the Eq.(2), and we calculated
defined in Eq.(3), meanwhile for EED and PL we used the
information obtained in the trace file. We use a tool developed
by Jaroslaw Malek named tracegraph to analyze the trace file
yielding [30],


B bits
T hroughput = 8
(2)
ttx
s
where, B is the number of transmitted bytes, and ttx is the
time of transmission in seconds. Also, parameter represents
the normalized throughput and is obtained as
=

T hroughput
RBR

where, RBR is the theoretical bite rate = 250 kbps.

(3)

Fig. 3. Normalized throughput as function of n in both: Tessellation and

Random scenarios, this metric had an indirect relationship related to the
number of nodes.

IV. S IMULATION R ESULTS

Figures 3, 4, and 5 show our main metrics related to the
number of nodes obtained in Tessellation and Randomly scenarios. Figure 3a shows an irregular decay of in Tessellation
scenario, its maximum and minimum values were 0.58 and
0.28 respectively; its boxplots corresponded a line because
its standard deviation was insignificant. Meanwhile, Fig. 3b
shows a directly relationship between and n in Randomly
scenario, its maximum and minimum values were 0.57 and
0.41 respectively; its boxplot showed a significant standard
deviation due to its own random nature and the number of
replications we have done.
The data suggest PL in both scenarios presented an increment as an exponential function of n, in Fig. 4a we determined
for Tessellation scenario in the range of nodes from 36 to 48
presented an irregularity, meanwhile in Fig. 4b we observed
Randomly scenario increase its PL directly to n, the main
difference between both scenarios was the Randomly scenario
had less PL than Tessellation scenario.
Finally, Fig. 5 shows the EED, we observed that in Fig.
5a Tessellation scenario presented more irregularity than Randomly scenario related to this metric, both scenarios presented
a mean value around to 3 ms.
We found based on the requirements gave from IGEPN
respect to PL must be less than 20%, Tessellation reached this

PL vs n

EED vs n

x 10

500
450

400

3.5

350

3
EndtoEnd

Paket Loss

300
250

2.5

200

1.5
150

100

0.5

50
0

12

18

24

30

36
42
nodes (n)

48

54

60

66

a) Tessellation Scenario

12

18

24

30

36
42
nodes (n)

48

54

60

66

48

54

60

66

a) Tessellation Scenario
PL vs n

EED vs n

x 10

500
450

400

3.5

350
3
EndtoEnd

Paket Loss

300
250
200

2.5

1.5
150
1

100

0.5

50
0

12

18

24

30

36
42
nodes (n)

48

54

60

66

b) Randomly Scenario

Fig. 4. Packet Loss as function of n in both: Tessellation and Random

scenarios, this metric had a direct relationship related to the number of nodes.

12

18

24

30

36
42
nodes (n)

b) Randomly Scenario

Fig. 5. End-to-end delay as function of n in both: Tessellation and Random

scenarios, this metric had a quasi-constant relationship related to the number
of nodes.
TABLE IV
C OMPARISON DEPLOYMENTS

value when n = 12 and Randomly reached this value when n =

18, Randomly scenario permit us to use more nodes with same
or better characteristics of and EED presented in Tessellation
scenario.

WSN
Deploy.

Variables/
Platform

Number
sensors

Frequency
Sampling
(Hz)
100
100
1000

Duration
(Days)

3
16
5

Operation
frequency
(MHz)
2400
2400
900

[31]
[32]
[33]
Proposed
Work

Aa /Micaz
SAb /Tmote
SA/Imote2
Sc /Micaz
e Iris

16

2400

100

2
19
Abiding

a Acustic
b Seismo-Acustic

V. R ESULTS FROM C OTOPAXI VOLCANO D EPLOYING

The Wireless Communications Research Group (WiCOM)

from Universidad de las Fuerzas Armadas ESPE, developed a
first attempt for replicating the predecessors works by using
Micaz and Iris platform. Thus, Cotopaxi, which is currently
the highest snowcapped volcano on Earth, was selected for
deploying a WSN. This work was deployed at an altitude of
4870 meters, where 16 sensor nodes were implemented, and
data were collected continuously for three days, Table. IV
summarize the comparison among previous works and our
deployment.

c Seismo

a) MICAz motes deployed on Volcano

Fig. 6.

b) IRIS motes deployed on Volcano

Wireless Sensor Networks deployed on Cotopaxi Volcano

One of the problems of this network is their dependence on

an aggregator, added to the information that must be stored

before being transmitted to the surveillance laboratory through
a wireless link of 40 km of distance where information is
processed in order to determine whether or not the events are
false alarms, this is the main problem to be solved, the signal
processing is off-line, because of that, it is not possible to
guarantee a real-time monitoring data [34].
Our first visit to the Cotopaxi Volcano was for locating
the geographic coordinates for placing the wireless communications system (0o 39 49 S, 78o 26 17 W), and to
determine the necessary requirements for WSN deployment on
Cotopaxi Volcano. In our second visit WSN were deployed,
at an altitude of 4870 meters, two differents WSNs, one
of them consisted of 10 motes MICAz, and another with 6
motes IRIS with MTS400 and MTS310 sensor cards using
two gateways MIB520 each. Mote Config 2.0 was the software
used to configure nodes. FTSP was implemented for time
synchronization. Fig. 6 shows the location of the two networks
deployed. Data were collected continuously for three days. The
energy problem was solved with a generator placed in situ. The
information was stored in a central station placed in situ, then
it was transmitted to the surveillance laboratory, located at a
distance of 40 km from volcano to ESPE, through a wireless
link.
A. WiFi-Based Long Distance Link
A wireless link was used for transporting data sensed by
WSN, it has proved to be cost effective for long distance
applications. The two major limitations for using WiFi over
long distances (WiLD) are the requirement for line of sight
between the endpoints and the vulnerability to interference in
the unlicensed band. Two further hurdles have to be overcome
when applying WiLD Technology, power budget and timing
limitations. The first was easily solved by using high gain
directional antennas, while the timing issue was addressed by
modifying the media access mechanism, as done by the TIER
group at the University of Berkeley [35].
For our purposes IEEE 802.11b was selected, basically,
because the 2,4 GHz ISM band presents less losses that
5,8 GHz ISM band, we used antennas gain of 24 dBi and
the transmission power of 1 watt. Meanwhile in the MAC
sublayer, three types of limitations can be extracted: the timer
waiting of ACKs, RTS/CTS and the definition of time related
to the slottime. We used Alix boards, in which we embedded
a middleware that permit us to modified these parameters in
order to link endpoints according. The performance of the link
was determined using DITG traffic injector, the injection time
was 1 minute, Fig. 7 is a portion of total file analyzed showing
that the mean throughput obtained is less than 2 Mbps and
a packet loss is less than 5%, this data rate is enough for
transmitting information sensed by our WSN, but the packet
loss must still be improved.
B. Seismic Signal Analysis
For a better representation of the signal is necessary to
have a representation in the time and frequency domain. Using
Fourier Transform, it is possible to detect the presence of a

a) Throughput

Fig. 7.

b) Number of packet loss

Parameters of Cotopaxi-ESPE link

certain frequency but it does not provide information about

the evolution over time of the spectral characteristics, for
this reason it is important to note that a spectrogram was
constructed using the Short Time Fourier Transform (STFT)
with a sampling frequency of 100 Hz.
The information in time and frequency domain of seismic
signals on a determined time often cannot be known, therefore,
we cannot define if the spectral component exists at any instant
of time, the only thing that can be distinguished are the time
intervals at certain frequency bands where we can find spectral
components.
Since seismic signals are non-stationary, it is necessary
to vary the size of the window, for this reason, we used
scale domain through wavelet analysis. The seismic signals
recorded was processed using discrete and continuous wavelet
transform, in order to discriminate the occurrence of a particular seismic event. The mother wavelet employed is DB6
(Daubechies 6), in the project were used decomposition levels
2 and 4 for the records of MICAz motes and IRIS motes
deployed respectively due to the amount of data they hold.
We must define if the event belongs to a volcanic tremor,
hybrid, long period or volcano tectonic events that permit us
to determine an abnormal behavior.
In Fig. 8, it was determined that the spectral components
present in the frequency range [1,25 - 2,5 Hz] are stable
and present a low energy, this signal corresponds to a noise
seismic event. Meanwhile, in Fig. 9, the signal spectral content
between [2,5 - 5 Hz] shows a seismic noise in a short period,
this is due to sudden temperature changes. In Fig. 10, the
signal between [5 - 10 Hz] is equally distributed in energy,
its acceleration remains constant during all time, this event
correspond to a volcanic tremor, this event is associated to
gas outlet due to high pressures. Finally Fig. 11 shows another
case, a high frequency volcanic tremor, signal between [10 20 Hz] it is related to strong gas outlet inside of the crater.
VI. C ONCLUSION AND F UTURE W ORKS
We determined the most optimal scenario in volcano monitoring is Randomly, it presents the best performance related
to all our metrics, to reach the PL 20% and the mean value
reached to EED = 3 ms, with its consequent of maximization
throughput. In this range we determine that the maximum
throughput is approximately equal to 145 kbps. This was
corroborated with an in-situ deployment, we determine a PL
equal to 25%, the mean value of EED = 1,1 s and a maximum
throughput equal to 130 kbps, the main difference between

Fig. 8.

Noise seismic event determined after digital processing signal

Fig. 10.

Volcanic tremor event determined after digital processing signal

Fig. 9. Short period noise seismic determined after digital processing signal

simulation and testbed was EED, data suggest this difference

due to the processing data must take to much time and this
values is unconsidered in simulation tools.
As future work we will modeling this system considering
the value of EED related to processing data, moreover we are
interested in feature extraction of volcano signals in order to
classify automatically these kind of events.
VII. ACKNOWLEDGMENTS
The authors gratefully acknowledge the contribution of
Universidad de las Fuerzas Armadas ESPE for the economical
support in the development of this project through the Wireless
Communications Research Group (WiCOM).
R EFERENCES
[1] R. Baheti and H. Gill, Cyber-physical systems, The Impact of Control
Technology, pp. 161166, 2011.
[2] R. Poovendran, Cyberphysical systems: Close encounters between two
parallel worlds [point of view], Proceedings of the IEEE, vol. 98, no. 8,
pp. 13631366, 2010.
[3] N. D. Stojanovic, L. Stojanovic, and R. Stuehmer, Tutorial: personal
big data management in the cyber-physical systems-the role of event
processing, in Proceedings of the 7th ACM international conference on
Distributed event-based systems. ACM, 2013, pp. 281288.
[4] L. Atzori, A. Iera, and G. Morabito, The internet of things: A survey,
Computer Networks, vol. 54, no. 15, pp. 27872805, 2010.
[5] J. P. Conti, The internet of things, Communications Engineer, vol. 4,
no. 6, pp. 2025, 2006.
[6] W. Shi and M. Liu, Tactics of handling data in internet of things,
in Cloud Computing and Intelligence Systems (CCIS), 2011 IEEE
International Conference on. IEEE, 2011, pp. 515517.

Fig. 11.
High frequency volcanic tremor event determined after digital
processing signal

[7] G. Werner-Allen, J. Johnson, M. Ruiz, J. Lees, and M. Welsh, Monitoring volcanic eruptions with a wireless sensor network, in Wireless
Sensor Networks, 2005. Proceeedings of the Second European Workshop
on. IEEE, 2005, pp. 108120.
[8] J. de Pedro Carracedo, Las tic en la prevencion de desastres naturales,
in II Congreso de Computacion para el Desarrollo (COMPDES09).
Acceso en: http://www. redusoi. org/docs/publicaciones/P11-Las 20TIC
20en% 20la% 20prevencion% 20de% 20desastres% 20nat urales. pdf,
2009.
[9] J. Zheng and A. Jamalipour, Wireless sensor networks: a networking
perspective. Wiley-IEEE Press, 2009.
[10] M. Kohvakka, M. Kuorilehto, M. Hannikainen, and T. D. Hamalainen,
Performance analysis of ieee 802.15. 4 and zigbee for large-scale
wireless sensor network applications, pp. 4857, 2006.
[11] I. Hammoodi, B. Stewart, A. Kocian, and S. McMeekin, A comprehensive performance study of opnet modeler for zigbee wireless sensor
networks, pp. 357362, 2009.
[12] Y. Li, C. S. Chen, Y.-Q. Song, Z. Wang et al., Real-time qos support
in wireless sensor networks: a survey, in 7th IFAC International Conference on Fieldbuses & Networks in Industrial & Embedded SystemsFeT2007, 2007.
[13] K. S. Prabh, Real-time wireless sensor networks, Ph.D. dissertation,
University OF Virginia, 2007.
[14] J. Gao, J. Hu, and G. Min, Performance modelling of ieee 802.15. 4
mac in lr-wpan with bursty on-off traffic, in Computer and Information
Technology, 2009. CIT09. Ninth IEEE International Conference on,
vol. 2. IEEE, 2009, pp. 5862.
[15] J. Gao, J. Hu, G. Min, and L. Xu, Qos performance analysis of ieee
802.15. 4 mac in lr-wpan with bursty error channels, in Mobile Ad-hoc
and Sensor Networks, 2009. MSN09. 5th International Conference on.
IEEE, 2009, pp. 252256.
[16] M. Imran, A. Said, and H. Hasbullah, A survey of simulators, emulators

[17]
[18]

[19]

[20]

[21]

[22]
[23]

[24]

[25]

[26]

[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]

and testbeds for wireless sensor networks, in Information Technology

(ITSim), 2010 International Symposium in, vol. 2, June, pp. 897902.
J. Zheng and M. J. Lee, A comprehensive performance study of ieee
802.15. 4, Sensor network operations, pp. 218237, 2004.
Y. Xue, H. S. Lee, M. Yang, P. Kumarawadu, H. H. Ghenniwa, and
W. Shen, Performance evaluation of ns-2 simulator for wireless sensor
networks, in Electrical and Computer Engineering, 2007. CCECE
2007. Canadian Conference on. IEEE, 2007, pp. 13721375.
F. Chen, N. Wang, R. German, and F. Dressler, Performance evaluation
of ieee 802.15. 4 lr-wpan for industrial applications, in Wireless on
Demand Network Systems and Services, 2008. WONS 2008. Fifth Annual
Conference on. IEEE, 2008, pp. 8996.
G. Tamilselvan and A. Shanmugam, Qualnet simulation of channel collision between ieee 802.15. 4 and ieee 802.11 b for various topologies,
International Journal of Engineering and Technology, vol. 1, no. 2, pp.
17938236, 2009.
D. Borade and S. Laeeq, Performance and evaluation of ieee 802.15.4
under different topologies with ad-hoc on demand distance vector
protocol, in Electrical, Electronics and Computer Science (SCEECS),
2012 IEEE Students Conference on, March, pp. 14.
J.-S. Lee, Performance evaluation of ieee 802.15.4 for low-rate wireless
personal area networks, Consumer Electronics, IEEE Transactions on,
vol. 52, no. 3, pp. 742749, 2006.
J.-S. Lee, Y.-M. Wang, and C.-C. Shen, Performance evaluation of
zigbee-based sensor networks using empirical measurements, in Cyber
Technology in Automation, Control, and Intelligent Systems (CYBER),
2012 IEEE International Conference on. IEEE, 2012, pp. 5863.
M. Petrova, J. Riihijarvi, P. Mahonen, and S. Labella, Performance
study of ieee 802.15. 4 using measurements and simulations, in Wireless
Communications and Networking Conference, 2006. WCNC 2006. IEEE,
vol. 1. IEEE, 2006, pp. 487492.
T. R. Burchfield, S. Venkatesan, and D. Weiner, Maximizing throughput
in zigbee wireless networks through analysis, simulations and implementations, in Proceedings of the International Workshop on Localized
Algorithms and Protocols for Wireless Sensor Networks Santa Fe, New
Mexico. Citeseer, 2007, pp. 1529.
M. Kohvakka, M. Kuorilehto, M. Hannikainen, and T. D. Hamalainen,
Performance analysis of ieee 802.15. 4 and zigbee for large-scale
wireless sensor network applications, in Proceedings of the 3rd ACM
international workshop on Performance evaluation of wireless ad hoc,
sensor and ubiquitous networks. ACM, 2006, pp. 4857.
G. Jimenez and M. Elena, Estudio de la eficiencia de encaminamiento
del protocolo aodv en redes ad hoc inalambricas de gran escala, 2009.
E. W. Weisstein, Schlafli symbol, from MathWorldA Wolfram Web
Resource. http://mathworld.wolfram.com/SchlaefliSymbol.html, visited
march 2013.
M. Stehlk, Comparison of simulators for wireless sensor networks,
2011.
K. N. J. Malek, Trace graph-data presentation system for network
simulator ns, in Tools and Applications (ISAT 2003), sep. 2003.
G. Werner-Allen, J. Johnson, M. Ruiz, J. Lees, and M. Welsh, Monitoring volcanic eruptions with a wireless sensor network, pp. 108120,
Jan 2005.
G. Werner-Allen, K. Lorincz, M. Ruiz, O. Marcillo, J. Johnson, J. Lees,
and M. Welsh, Deploying a wireless sensor network on an active
volcano, Internet Computing, IEEE, vol. 10, no. 2, pp. 1825, 2006.
G. Werner-Allen, K. Lorincz, J. Johnson, J. Lees, and M. Welsh,
Fidelity and yield in a volcano monitoring sensor network, pp. 381
396, 2006.
I. F. Londono Mancheno, Implementacion de un sistema de monitorizacion de senales ssmicas del volcan cotopaxi empleando una red de
sensores inalambricos, 2011.
R. K. Patra, S. Nedevschi, S. Surana, A. Sheth, L. Subramanian, and
E. A. Brewer, Wildnet: Design and implementation of high performance
wifi based long distance networks. vol. 1, no. 1, p. 1, 2007.