ArticlePDF Available

Selecting the Best Mobility Model with the AODV Routing Protocol in MANETs

Authors:

Abstract and Figures

A mobile ad hoc network (MANET) is an autonomous, self-configuring network of mobile nodes that can be formed without the need of any pre-established infrastructure or centralized administration. MANETs are extremely flexible and each node is free to move independently, in any random direction. Each node in the MANET maintains continuously the information required to properly route traffic. This paper presents a simulation study to analyze and evaluate the behavior of the MANET with AODV routing protocol by testing four mobility models (i.e. Waypoint(RWP), Reference Point Group Model (RPGM), Gauss Markov Model (GMM) and Manhattan Grid Model (MGM)). Several performance metrics (Throughput, Packet Delivery Fraction (PDF), Average End-to-end Delay (AED),Normalize Routing Load (NRL) and packets loss) were suggested as a measuring tool to be used in the comparison stage for all these four mobility models using NS-2.Various parameters such as different number of nodes, different speeds, different pause times, different environment areas and different traffic rates were also used in five suggested scenarios. The results indicated that the best performance of AODV routing protocol is with RPGM mobility model than other mobility models
Content may be subject to copyright.
ORIENTORIENT
ORIENTORIENT
ORIENTAL JOURNAL OFAL JOURNAL OF
AL JOURNAL OFAL JOURNAL OF
AL JOURNAL OF
COMPUTER SCIENCE & TECHNOLCOMPUTER SCIENCE & TECHNOL
COMPUTER SCIENCE & TECHNOLCOMPUTER SCIENCE & TECHNOL
COMPUTER SCIENCE & TECHNOLOGOG
OGOG
OGYY
YY
Y
www.computerscijournal.org
ISSN: 0974-6471
June 2013,
Vol. 6, No. (2):
Pgs.75-84
An International Open Free Access, Peer Reviewed Research Journal
Published By:
Oriental Scientific Publishing Co., India.Oriental Scientific Publishing Co., India.
Oriental Scientific Publishing Co., India.Oriental Scientific Publishing Co., India.
Oriental Scientific Publishing Co., India.
INTRODUCTIONINTRODUCTION
INTRODUCTIONINTRODUCTION
INTRODUCTION
The Mobile Ad-hoc Network (MANET) is
a collection of nodes, which have the possibility
of connecting in a wireless medium forming an
arbitrary dynamic network. Such mobile network
can dynamically change with time, new nodes can
join, and other nodes can leave the network [C.P.
Agrawal et al.2008]. A large majority of MANET
studies are based on simulating the Random
Waypoint mobility models, which is one of the
default cases in the Network Simulator (NS-2). In
Selecting the Best Mobility Model with the ASelecting the Best Mobility Model with the A
Selecting the Best Mobility Model with the ASelecting the Best Mobility Model with the A
Selecting the Best Mobility Model with the AODOD
ODOD
ODV RV R
V RV R
V Routingouting
outingouting
outing
Protocol in MANETsProtocol in MANETs
Protocol in MANETsProtocol in MANETs
Protocol in MANETs
SASA
SASA
SAAD TALIB HASSON AD TALIB HASSON
AD TALIB HASSON AD TALIB HASSON
AD TALIB HASSON andand
andand
and
ALAL
ALAL
ALAA
AA
AA TAIMAA TAIMA
A TAIMAA TAIMA
A TAIMA
Computer Sciences Department, University of Babylon, (Iraq).
Corresponding author E-mail :saad_aljebori@yahoo.com,alaa_alhasany1983@yahoo.com
(Received: May 20, 2013; Accepted: June 02, 2013)
ABSTRACTABSTRACT
ABSTRACTABSTRACT
ABSTRACT
A mobile ad hoc network (MANET) is an autonomous, self-configuring network of mobile
nodes that can be formed without the need of any pre-established infrastructure or centralized
administration. MANETs are extremely flexible and each node is free to move independently, in
any random direction. Each node in the MANET maintains continuously the information required to
properly route traffic. This paper presents a simulation study to analyze and evaluate the behavior
of the MANET with AODV routing protocol by testing four mobility models (i.e. Waypoint(RWP),
Reference Point Group Model (RPGM), Gauss Markov Model (GMM) and Manhattan Grid Model
(MGM)).
Several performance metrics (Throughput, Packet Delivery Fraction (PDF), Average
End-to-end Delay (AED),Normalize Routing Load (NRL) and packets loss) were suggested as a
measuring tool to be used in the comparison stage for all these four mobility models using NS-
2.Various parameters such as different number of nodes, different speeds, different pause times,
different environment areas and different traffic rates were also used in five suggested scenarios.
The results indicated that the best performance of AODV routing protocol is with RPGM mobility
model than other mobility models.
KK
KK
Key words : ey words :
ey words : ey words :
ey words : MANET , AODV , RWP, RPGM , GMM , MGM , Performance evaluation.
the last years, different mobility models have been
proposed, with the goal of reproducing realistic
node movement as one of the major concerns [C.
Gomez et al.2004]. It is so important to first
understand and evaluate the performance of the
available routing protocols in different mobility
scenarios before selecting a most suitable protocol
for any particular scenario. Most previous studies
with routing protocols selected the Random
Waypoint mobility model for simulations. However,
surveys on mobility models and impact on routing
performance verify that the analysis of the protocol
76
performance using just Random Waypoint model
is not enough; a given routing protocol may not
deliver optimum performance under other mobility
models [Fahim Maan et al.2011].
Related workRelated work
Related workRelated work
Related work
R. Manoharan, et al. at 2010 studied three
widely used mobility models such as Random Way
Point, Reference Point Group and Manhattan
mobility that in addition to the strengths and
weaknesses of the individual multicast routing
protocols, the mobility patterns does also have
influence on the performance of the routing
protocols. Multicast Ad hoc On-demand Distance
Vector Routing protocol and adaptive demand
driven multicast routing protocol have been chosen
and implemented in NS2. They observed that the
mobility patterns do also have influence on the
performance of the routing protocols [R.
Manoharan , et al. 2010].
Sunil Kumar Kaushik et al. at 2012
analyzed the behavior of five MANETs routing
protocols (i.e. AODV, DSDV, DSR, OLSR and TORA)
under the three mobility models (RPGM, CMM and
RWP) and then they compared the performance
of these protocols using NS-2 simulator in certain
area of (700 x 700 m2). These routing protocols
were compared in their; (PDR), (delay), (NRL) and
Throughput with the change in numbers of nodes.
Their simulation results showed that the Reactive
protocols is much better than the proactive in the
packet delivery (PDR), End-to-End delay (Delay),
Normalized routing load (NRD) and throughput
[Sunil Kumar Kaushik and et al. 2012].
Prajakta M. Dhamanskar, et al. at 2012
presented the performance of on-demand routing
protocols such as AODV, DSR and TORA for mobile
nodes following four mobility models such as
Random Waypoint (RWP), Random Walk (RW),
Manhattan Grid and Reference Point Group
mobility model (RPGM). They stated from their
simulation results that the performance of RPGM
mobility model is the best and the performance of
Manhattan Grid is the worst as compared to other
mobility models for all these three protocols. PDR
of AODV and TORA is greater than that of DSR but
PDR of TORA is the best. NRL is the least in DSR
and Delay is average in TORA [Prajakta M.
Dhamanskar and et al. 2012].
Routing protocolsRouting protocols
Routing protocolsRouting protocols
Routing protocols
Numbers of routing protocols for Ad Hoc
networks were developed and used. Protocols
were classified as proactive and reactive protocols
[Ejiro .E. Igbesoko et al.2010]. This work focuses
on applying and using the AODV as a reactive
protocol. AODV Protocol stands for Ad-hoc On-
Demand Distance Vector Routing which maintains
a routing table at each node. It is proactive type &
contains three essential entries in the routing table
for a destination, a next hop node, a sequence
number and a hop count. All packets directed to
the destination are sent to the next hop node. The
sequence number measures the freshness of a
the route. The hop count represents the current
distance to the destination node [C. P. Agrawal et
al., 2008].
Mobility ModelsMobility Models
Mobility ModelsMobility Models
Mobility Models
A mobility model should be attempted to
emulate the movements of the real mobile nodes.
Mobility models are based on setting out different
parameters related to the possible nodes
movement. The basic parameters are the starting
location of mobile nodes, their movement direction,
velocity range, and speed changes over time.
Mobility models can be classified into entity and
group models. Entity models covers scenarios
when mobile nodes move completely
independently from each other, while in group
models nodes are dependent on each other or on
some predefined leader node [T. Camp et al.2002].
In this paper, the following Mobility models were
studied:
RR
RR
Random Wandom W
andom Wandom W
andom Waypoint Mobility Model (RWP)aypoint Mobility Model (RWP)
aypoint Mobility Model (RWP)aypoint Mobility Model (RWP)
aypoint Mobility Model (RWP)
In RWP mobility model, each node of the
network selects a random destination and moves
towards it with certain chosen random velocity.
Once a node reaches the destination, the node
stops for a duration defined by the pause time
parameter. After pause time duration, node again
selects a random destination and repeats the
whole process again until the simulation ends [K.
Amjad, et al. 2010].
Reference Point Group Mobility Model (RPGM)Reference Point Group Mobility Model (RPGM)
Reference Point Group Mobility Model (RPGM)Reference Point Group Mobility Model (RPGM)
Reference Point Group Mobility Model (RPGM)
In reference point group mobility model,
77
nodes are divided into groups. Every group has a
group leader that determines the movements of
all nodes in the group. At each instant, speed and
direction of group member is calculated based on
speed and direction of leader node at that instant.
This model represents the movement of soldiers
in a battalion, or tourists following a tourist guides
[Sri Chusri Haryanti, et al. 2011].
Gauss-Markov Mobility ModelGauss-Markov Mobility Model
Gauss-Markov Mobility ModelGauss-Markov Mobility Model
Gauss-Markov Mobility Model
In this model, initially each mobile node
is assigned a current speed and direction at each
fixed interval of time. Node movement occurs by
updating the speed and direction of each mobile
node. Because of temporal dependency, the value
of speed and direction at the particular time is
calculated on the basis of the value of previous
speed and direction. This model eliminates the
abrupt stops, quick turns and is close to be realistic
[Valentina Timcenko, et al. 2010].
Manhattan Mobility ModelManhattan Mobility Model
Manhattan Mobility ModelManhattan Mobility Model
Manhattan Mobility Model
In Manhattan model, movement pattern
of mobile nodes were defined by map which
composed of a number of horizontal and vertical
streets. Node allows moving along the grid of the
horizontal and vertical streets on the map. Because
of temporal dependency, velocity of a mobile node
at a particular time is dependent on the velocity of
its previous time [Krunal Ptel, et al. 2012] .
Network Simulator NS-2Network Simulator NS-2
Network Simulator NS-2Network Simulator NS-2
Network Simulator NS-2
The network simulator NS-2 is a discrete
event simulation software for network simulations.
It simulates events such as receiving, sending,
dropping and forwarding packets. NS-2.34 can be
built on different platforms. It also offers a visual
representation of the simulated network by tracing
nodes events, movements and writing them in a
file called a Network animator (NAM file) [Neha
Rani, et al.2012].
This simulation study has been done
using the NS-2 as a network simulator. A Linux
platform (Ubuntu) was chosen. Linux offers a
number of programming development tools that
can be used with the required simulation process.
Performance metricsPerformance metrics
Performance metricsPerformance metrics
Performance metrics
In this simulation study the (Throughput,
Packet delivery fraction, Average end-to-end delay,
Normalized Routing Load and Packets Loss) were
used as the main performance metrics indicators
to evaluate , analyze and compare the network
behaviors with each mobility model scenario.
Methodology for Performance EvaluationMethodology for Performance Evaluation
Methodology for Performance EvaluationMethodology for Performance Evaluation
Methodology for Performance Evaluation
The following steps were suggested in
this paper to evaluate the impact of the mobility
models on the performance of 5 Metrics for an
AODV routing protocol in MANET :
Step 1Step 1
Step 1Step 1
Step 1
Start.
Step 2Step 2
Step 2Step 2
Step 2
Create the traffic generation file "CBR file"
that generated by cbrgen.tcl (this script found in
ns-allinone-2.34/ns-2.34/ind_util/cmu_scen_gen/ ).
Step 3Step 3
Step 3Step 3
Step 3
Set p = 0 (this variable to determine the
number of evaluation cases(parameters)).
Step 4Step 4
Step 4Step 4
Step 4
select the parameters ( evaluation cases).
this simulation includes varying number of nodes,
varying speeds , varying areas, vary pause times
and varying traffic rates.
Step 5Step 5
Step 5Step 5
Step 5
set i=0 ( this variable to determine the no.
of mobility models).
Step 6Step 6
Step 6Step 6
Step 6
select the mobility models which used to
determine to describe the movement pattern of
mobile users, and how their location, velocity and
acceleration change over time. This paper includes
random waypoint, reference point group , gauss
markov and manhattan grid model. This mobility
models will be generated by setdest or by
BonnMotion .
Step 7Step 7
Step 7Step 7
Step 7
set s=0 (no. of scenario file (movement file) ) .
Step 8Step 8
Step 8Step 8
Step 8
select the scenario file which used to
determine the no. of nodes , nodes speed, pause
times , simulation time and dimension of the
topography for each mobility model.
Step 9Step 9
Step 9
Step 9
Step 9
create tcl file that represent simulation
environment of MANET with mobility model for one
routing protocol.
78
Step 10Step 10
Step 10Step 10
Step 10
add tcl file as input into NS-2 in order to
perform the simulation , the output are NAM and
Trace file.
Step 11Step 11
Step 11Step 11
Step 11
using NAM file to display all event through
the simulation as visualization review , while the
trace file will be used to compute the performance
metrics (such as throughput, packet delivery
fraction, average end-to-end delay , NRL and no.
of packets loss) using AWK programming
language.
Step12Step12
Step12Step12
Step12
Increment s by 1.
Step 13Step 13
Step 13Step 13
Step 13
if ( s<10) then go to step 8 (s is no. of
scenario files is 10).Otherwise go to step 14.
Step 14Step 14
Step 14Step 14
Step 14
Increment i by 1.
Step 15Step 15
Step 15Step 15
Step 15
If (i < 4) then go to step 6 (i is no. of mobility
models evaluated in this study ) . Otherwise , go to step 16.
Step 16Step 16
Step 16Step 16
Step 16
Increment p by 1.
Step 17Step 17
Step 17Step 17
Step 17
If ( p < 5) then go to step 4. (p is no. of
evaluation parameters) Otherwise go to step 18
Step 18Step 18
Step 18Step 18
Step 18
split the result files into no. of files (the
number of files depends on number mobility
models that will be evaluated in this paper).
Step 19Step 19
Step 19Step 19
Step 19
calculate the final average of
performance metrics for all mobility models that
will be evaluated to represent its impact on MANET's
performance .
Step 20Step 20
Step 20Step 20
Step 20
split the final average from previous step
into no. of files which used to draw the result.
Step 21Step 21
Step 21Step 21
Step 21
draw the results by Xgraph , TraceGraph
or by excel.
Step22:Step22:
Step22:Step22:
Step22: End.
The following flow chart shown in Fig 1
clarifies the implementation stages of the proposed
system for the performance evaluation :
Simulation EnvironmentSimulation Environment
Simulation EnvironmentSimulation Environment
Simulation Environment
This simulation study was implemented
TT
TT
Table. 2 simulation environment
able. 2 simulation environment
able. 2 simulation environmentable. 2 simulation environment
able. 2 simulation environment
Network Simulator Network Simulator
Network Simulator Network Simulator
Network Simulator
The simulator NS-2.34
NA M 1.13
MAC Type 802.11
Radio Propagation Two Ray
Model Ground
Antenna Type Omni Antenna
T T
T T
Traffic and Mobilityraffic and Mobility
raffic and Mobilityraffic and Mobility
raffic and Mobility
Data Traffic Type C B R
Simulation Time 75 second
Data Payload 512 bytes
Interface Queue Type Drop Tail /
Pri Queue
Mobility Models RWP , RPGM ,
GMM and MGM
Routing Protocols Routing Protocols
Routing Protocols Routing Protocols
Routing Protocols
Routing Protocols AODV
on personal computer with Pentium core2due
processor, 2.4 GHz CPU, 2 GB RAM, 320 GB Hard
Disk and Linux - Ubuntu 10.10 Operating System.
Table (1) presents the suggested MANET's
simulation environment implemented in this paper.
Fig .1: Implementation stages of the proposedFig .1: Implementation stages of the proposed
Fig .1: Implementation stages of the proposedFig .1: Implementation stages of the proposed
Fig .1: Implementation stages of the proposed
system for the performance evaluation.system for the performance evaluation.
system for the performance evaluation.system for the performance evaluation.
system for the performance evaluation.
79
Simulation ResultsSimulation Results
Simulation ResultsSimulation Results
Simulation Results
In this section, five scenarios were suggested
and implemented to evaluate and analyze the
performance of mobility models for MANET, these
parameters determine the impact of mobility
models on the performance of MANET routing
protocols. These parameters will be investigated
as shown in Table 3 .
The simulation carried out 10 times for
each mobility model, the sum of times is 40 for the
four mobility models, the total number of times is
160 for all mobility models under five parameters.
The performance metrics used in this evaluation
study are; packet delivery fraction (PDF),
throughput, no. of lost packets, normalized routing
load (NRL) and average end-to-end delay (AED).
The main used parameters in this paper are varying
no. of nodes, varying speeds, varying pause times
, varying simulation area and varying traffic rates.
The results are shown in the following Fig.
Fig 2 shows the behavior of 5 AODV
performance metrics (i.e. Throughput, PDF, Packets
loss, NRL and AED) under four mobility models
(i.e. RWP, RPGM, GMM and MGM) in the first
Scenario (varying number of nodes).The
throughput of AODV is more significant with RPGM
and RWP and it is less significant with GMM and
MGM.The PDF of AODV is best in RPGM and RWP.
PDF in GMM and MGM is less than the others.NRL
is decreased when the no. of nodes increased.
NRL in RPGM is low because the group leader
decides the speed of the group members. In MGM
NRL is high. The no. of packets loss in GMM and
MGM is higher than RPGM and RWP, the packets
loss are increased when the no. of nodes
decreased. The AED is decreased when the no. of
nodes increased. AED in RPGM is the least and in
GMM and with MGM is highest
Fig 3 shows the behavior of 5 AODV
performance metrics (i.e. Throughput, PDF, Packets
loss, NRL and AED) under four mobility models
(i.e. RWP,RPGM,GMM and MGM) with Scenario 2
(varying nodes speeds).The throughput of AODV
was decreased when the nodes speed were
increased . RPGM and RWP have high throughput
and while MGM and GMM having the lowest
values.The PDF of AODV were decreased when
the node speed were increased . RPGM and RWP
have high PDF while MGM and GMM having less
values. With all the Mobility models, the PDF values
increased to certain level (with speed = 20).The
no. of the lost packets in GMM and MGM is highest
while in RPGM and RWP is lowest, the loss packet
are increased when the node speed increased
after nodes speed exceeds 20.The NRL of this
The AED is increased when the node speed
increased . AED in RPGM is the lowest and with
MGM and GMM is highest.protocol is increases
with high speed for all mobility models. RPGM has
low NRL than other mobility models while MGM
has high NRL The AED is increased when the
node speed increased . AED in RPGM is the lowest
and with MGM and GMM is highest.
TT
TT
Table . 3 : General Pable . 3 : General P
able . 3 : General Pable . 3 : General P
able . 3 : General Parameters for All scenariosarameters for All scenarios
arameters for All scenariosarameters for All scenarios
arameters for All scenarios
No.No.
No.No.
No. Scenario NameScenario Name
Scenario NameScenario Name
Scenario Name No. of nodesNo. of nodes
No. of nodesNo. of nodes
No. of nodes Node SpeedNode Speed
Node SpeedNode Speed
Node Speed PP
PP
Pause Timeause Time
ause Timeause Time
ause Time Area SizeArea Size
Area SizeArea Size
Area Size TT
TT
Trafficraffic
rafficraffic
raffic
RateRate
RateRate
Rate
1 No. of Nodes 25 , 50
75,100 20 15 1000*1000 4
2 Node Speeds 25 10 , 20
40 , 60 10 1000*1000 4
3 Pause Times 50 40 0 , 6
10 ,14 1000*1000 4
4 Area Sizes 60 20 12 500*500 , 700*700
1000*1000 , 4
1200*1200
5 Traffic Rates 75 15 10 1000*1000 4,8
12, 16
80
FF
FF
Fig 2[a-e]: The behavior of 5 Aig 2[a-e]: The behavior of 5 A
ig 2[a-e]: The behavior of 5 Aig 2[a-e]: The behavior of 5 A
ig 2[a-e]: The behavior of 5 AODOD
ODOD
ODVV
VV
V
performance metrics under four mobilityperformance metrics under four mobility
performance metrics under four mobilityperformance metrics under four mobility
performance metrics under four mobility
models with varying number of nodes.models with varying number of nodes.
models with varying number of nodes.models with varying number of nodes.
models with varying number of nodes.
FF
FF
Fig 3[a-e]: The behavior of 5 Aig 3[a-e]: The behavior of 5 A
ig 3[a-e]: The behavior of 5 Aig 3[a-e]: The behavior of 5 A
ig 3[a-e]: The behavior of 5 AODOD
ODOD
ODVV
VV
V
performance metrics under four mobilityperformance metrics under four mobility
performance metrics under four mobilityperformance metrics under four mobility
performance metrics under four mobility
models with varying nodes speed.models with varying nodes speed.
models with varying nodes speed.models with varying nodes speed.
models with varying nodes speed.
81
FF
FF
Fig 5 [a-e]: The behavior of 5 Aig 5 [a-e]: The behavior of 5 A
ig 5 [a-e]: The behavior of 5 Aig 5 [a-e]: The behavior of 5 A
ig 5 [a-e]: The behavior of 5 AODOD
ODOD
ODVV
VV
V
performance metrics under four mobilityperformance metrics under four mobility
performance metrics under four mobilityperformance metrics under four mobility
performance metrics under four mobility
models with varying areas.models with varying areas.
models with varying areas.models with varying areas.
models with varying areas.
FF
FF
Fig 4[a-e]: The behavior of 5 Aig 4[a-e]: The behavior of 5 A
ig 4[a-e]: The behavior of 5 Aig 4[a-e]: The behavior of 5 A
ig 4[a-e]: The behavior of 5 AODOD
ODOD
ODVV
VV
V
performance metrics under four mobilityperformance metrics under four mobility
performance metrics under four mobilityperformance metrics under four mobility
performance metrics under four mobility
models with varying pause times.models with varying pause times.
models with varying pause times.models with varying pause times.
models with varying pause times.
82
Fig 4 shows the behavior of 5 AODV
performance metrics (i.e. Throughput, PDF, Packets
loss, NRL and AED) under four mobility models
(i.e. RWP, RPGM, GMM and MGM) with varying
pause times. Throughput of RPGM is extremely
better than all the other mobility models and MGM
and GMM have clearly worst results. The AODV
has best PDF with RPGM mobility model . RWP is
better next to RPGM. PDF in MGM and GMM is
very low in comparison of RPGM RWP mobility
Models. In MGM and GMM, the no. of packets
loss increased when the value of the pause times
increased, RPGM and RWP provides a lowest no.
of packet loss. The normalized routing load of
AODV can be easily arranged in an order from
best to worst as follows: RPGM, RWP, GMM and
MGM.The RPGM and KWP exhibit the lowest delay
and GMM and MGM highest delay.
Fig 5 shows the behavior of 5 AODV
performance metrics (i.e. Throughput, PDF, Packets
loss, NRL and AED) under four mobility models
(i.e. RWP, RPGM, GMM and MGM) with varying
simulation areas.The throughput of AODV became
highest in RPGM and RWP and is lowest with MGM
and GMM.The PDF of AODV became highest in
RPGM and RWP and is lowest with MGM and
GMM.The no. of packets loss in GMM and MGM is
highest while in RPGM and RWP is lowest. The
packets loss are increased when the environment
size increased,The NRL of this protocol is
decreases with large environment size for all
mobility models. RPGM has low NRL than other
mobility models while MGM has the higher
value.The AED is increased when the environment
size is increased . AED in RPGM is lowest and in
MGM and GMM is the highest
Fig 6 shows the behavior of 5 AODV
performance metrics (i.e. Throughput, PDF, Packets
loss, NRL and AED) under four mobility models
(i.e. RWP, RPGM, GMM and MGM) with varying
traffic rates.The throughput of AODV became lower
when the network load is higher. This protocol is
highest throughput with RPGM and lowest with
MGM and GMM.The PDF of AODV became lower
when the network load is higher. This protocol is
highest PDF with RPGM and lowest with MGM
and GMM.The no. of packets loss in GMM and
MGM is highest while in RPGM and RWP is lowest,
FF
FF
Fig 6[a-e]: The behavior of 5 Aig 6[a-e]: The behavior of 5 A
ig 6[a-e]: The behavior of 5 Aig 6[a-e]: The behavior of 5 A
ig 6[a-e]: The behavior of 5 AODOD
ODOD
ODVV
VV
V
performance metrics under four mobilityperformance metrics under four mobility
performance metrics under four mobilityperformance metrics under four mobility
performance metrics under four mobility
models with varying traffic rates.models with varying traffic rates.
models with varying traffic rates.models with varying traffic rates.
models with varying traffic rates.
83
the loss packet are increased when the traffic rate
increased.The NRL is decreased when the traffic
rate is increased. The NRL in RPGM is the lowest
and in MGM is the highest .The AED of AODV is
increased when the traffic rate increased . This
protocol with GMM and MGM shows highest AED
but with RPGM gives the lowest AED values.
CONCLUSIONCONCLUSION
CONCLUSIONCONCLUSION
CONCLUSION
In this paper, the performance of the four
mobility models was evaluated and analyzed
using NS-2 and Bonn Motion according to 5
performance metrics . This evaluation study shows
that the RPGM was best mobility model suited for
AODV routing protocol when compared to other
available mobility models.
1. [C. Gomez et al. 2004]. C. Gomez, X.
Marchador, V. Gonzalez and J. Paradells,
“Multilayer analysis of the influence of
mobility models on TCP flows in AODV ad-
hoc networks”, FEDER and the Spanish
Government through project TIC2003-
01748, Spain, (2005).
2. [C.P.Agrawal et al. 2008]. C.P.Agrawal, M.K
Tiwari, and O.P.Vyas, “Evaluation of AODV
Protocol for Varying Mobility Models of
MANET for Ubiquitous Computing”, IEEE,
Third 2008 International Conference on
Convergence and Hybrid Information
Technology, India., (2008).
3. [Ejiro .E. Igbesoko et al. 2010] Ejiro .E.
Igbesoko, Thaddeus Onyinye Eze and
Mona Ghassemian, “ Performance Analysis
of MANET Routing Protocols over Different
Mobility Models “, London, (2009).
4. [Fahim Maan et al. 2011]. Fahim Maan and
Nauman Mazhar, “MANET Routing
Protocols vs Mobility Models: A
Performance Evaluation”, IEEE, (2011).
5. [K. Amjad et al. 2010] K. Amjad and A.J.
Stocker, “Impact of node density and
mobility on the performance of AODV and
DSR in MANETS” , IEEE, (2010).
6. [Krunal Ptel et al. 2012] Krunal Ptel and
Krunal Ptel, “Performance Evaluation of
Stable AODV Routing Protocol under
Different Mobility Models”,
International
Journal of Computer Applications
(0975 –
8887) 43(8)43(8)
43(8)43(8)
43(8), (2012).
7. [Neha Rani et al. 2012] Neha Rani, Preeti
Sharma and Pankaj Sharma, “Performance
Comparison Of Various Routing Protocols
REFERENCESREFERENCES
REFERENCESREFERENCES
REFERENCES
Indifferent Mobility Models” ,
International
Journal of Ad hoc, Sensor & Ubiquitous
Computing (IJASUC)
3(4)3(4)
3(4)3(4)
3(4), (2012) .
8. [Prajakta M. Dhamanskar and et al. 2012]
Prajakta M. Dhamanskar and Dr. Nupur Giri,
“Performance evaluation of On Demand
MANET protocols for different mobility
models”,
International Journal of Emerging
Technology and Advanced Engineering
,
ISSN 2250-2459, 2(9)2(9)
2(9)2(9)
2(9) ,(2012).
9. [R. Manoharan et al.2010] R. Manoharan
and E. Ilavarasan, “IMPACT OF MOBILITY
ON THE PERFORMANCE OF MULTICAST
ROUTING PROTOCOLS IN MANET”,
International Journal of Wireless & Mobile
Networks(IJWMN), 2(2)2(2)
2(2)2(2)
2(2),(2010).
10. [Sri Chusri Haryanti et al. 2011] Sri Chusri
Haryanti and Riri Fitri Sari, “Effect of RPGM
and Gauss Markov Node Mobility for
Resource Allocation Performance in Mobile
Ad Hoc Computational Grid”,
IJCSNS
International Journal of Computer Science
and Network Security
, 11(6)11(6)
11(6)11(6)
11(6), (2011).
11. [Sunil Kumar Kaushik et al. 2012] Sunil
Kumar Kaushik and Sukhbir Singh ,”
Performance Evaluation of Mobile Ad Hoc
Networks with Reactive and Proactive
Routing Protocols and Mobility Models “ ,
International Journal of Computer
Applications
(0975 – 8887), 54(17)54(17)
54(17)54(17)
54(17), (2012).
12. [T. Camp et al. 2002]. Tracy Camp, Jeff
Boleng and Vanessa Davies “A Survey of
Mobility Models for Ad Hoc Network
Research”,
Wireless Communication &
Mobile Computing (WCMC)
: 2(5)2(5)
2(5)2(5)
2(5), pp. 483-
502, (2002).
84
13. [Valentina Timcenko et al.2010] Valentina
Timcenko, Mirjana Stojanovic and Slavica
Bostjancic Rakas, “MANET Routing
protocols vs. mobility models: Performance
analysis and comparison” Proceedings of
the 9th WSEAS International Conference
on Applied Informatics and
Communications (AIC ’09), (2010).
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
Mobile Ad hoc Network (MANET) is a infrastructure l ess network in which two or more devices have wireless communication which can communicate with e ach other and exchange information without need of any centralized administrator. Each node in the ad hoc network acts as a router, forwarding data packets for other nodes. The main issue is to compa re the existing routing protocol and finding the be st one. The scope of this study is to test routing per formance of three different routing protocols (AODV , OLSR and DSDV) with respect to various mobility mod els using NS2 simulator. In this paper the parameters used for comparison are packet delivery fraction (PDF), average end to end delay (AEED), normalized routing load (NRL) and throughput
Article
Full-text available
Mobile ad hoc computational grid enables resource sharing between mobile nodes in an ad hoc networks for completing specific tasks. To maximize utilization of sharing resources in such grid, an effective resource allocation algorithm that is suitable with the environment, plays a key role. In this paper, we investigate the performance of resource allocation scheme EERA that has been designed for mobile ad hoc computational grid in RPGM and Gauss-Markov node mobility model. We perform the simulation in NS2. The result shows that accumulative application completion times for the scheme using Gauss-Markov mobility nodes is 45% up to 350% greater than using RPGM model. Increasing the maximum node's speed will increase the accumulative application completion times. Meanwhile the average end-to-end delay and the number of drop packet for Gauss-Markov is lower compare to RPGM in maximum speed of 5m/s.
Article
Full-text available
Mobile Ad hoc Network is a collection of wireless mobile nodes dynamically forming a temporary network without the aid of any established infrastructure or centralized administration. Routing protocols in mobile ad hoc network helps node to send and receive packets. In this paper we are doing study of AODV, DSR (Reactive), and OLSR, DSDV, TORA (Proactive) protocols based on various mobility models [3] such as RPGM, CMM and RWP . In this paper we evaluate performance of five types of routing protocols (AODV, DSR, OLSR, DSDV and TORA) based on packet delivery ratio, average end to end delay, routing overhead and throughput. In this paper we will analyze and compare the performance of reactive and proactive routing protocols under different mobility models using NS-2 simulator in the area of 700 x 700 m 2 .
Article
Full-text available
This work is a simulation based comparison of three Mobile Ad Hoc Networks (MANETs) routing protocols – Destination Sequenced Distance Vector (DSDV), Ad hoc on demand Distance Vector (AODV) and Dynamic Source Routing (DSR). The comparison is based on their performance in some existing mobility models used in the simulation of MANETs. The mobility models used in this work are Random Waypoint, Manhattan Grid, Gauss-Markov, Reference Point Group and Heterogeneous Mobility Models. With detailed simulations of the protocols across these models, comparisons are made based on results obtained using known metrics for evaluating the protocols. Also, the protocols are compared under varied categories of disaster area scenarios modelled by a Heterogeneous mobility model. Across the models with respect to considered metrics for comparison, DSR outperformed others followed by AODV and then DSDV.
Article
Mobile Ad hoc Network (MANET) is a infrastructure less network in which two or more devices have wireless communication which can communicate with each other and exchange information without need of any centralized administrator. Each node in the ad hoc network acts as a router, forwarding data packets for other nodes. The main issue is to compare the existing routing protocol and finding the best one. The scope of this study is to test routing performance of three different routing protocols (AODV, OLSR and DSDV) with respect to various mobility models using NS2 simulator. In this paper the parameters used for comparison are packet delivery fraction (PDF), average end to end delay (AEED), normalized routing load (NRL) and throughput.
Article
BSTRACTAd hoc Networks are infrastructure less networks. Nodes themselves do routing and forwarding functions. Topology changes due to node movement. Frequent changes in topology leads to more route breaks. This in turn leads to reduction in Packet delivery Ratio (PDR). It is desired that routing protocols should select stable paths i.e. which are less likely to break. In this work, we have implemented one such protocol. Different mobility patterns lead to different performance of routing protocols. We have evaluated stable routing protocol under following mobility models: Random WayPoint, Manhattan Model, Reference Point Group Mobility and Gauss Markov Model. Performance measures of interest are Packet Delivery Ratio (PDR) and routing overhead. It is found that RPGM results in better PDR and lowest routing overhead compared to other models. Manhattan model results in lowest PDR and highest routing overhead.
Article
The fundamental characteristic which differentiates MANETs from other wireless or wired networks is mobility. Therefore, MANET routing protocols are designed to adaptively cater for dynamic changes in topology while maximizing through- put and packet delivery ratio, and minimizing delay, routing load and energy consumption. A major design issue for an efficient and effective routing protocol for real MANETs is, therefore, to achieve optimum values of performance parameters under network scenarios where nodes are subjected to different types of mobility that dynamically change the network topology. Our sim- ulative study on MANET routing protocols and mobility models aims to determine the performance of current MANET routing protocols with respect to various mobility models implemented in ns-2. We compare a number of reactive and proactive routing protocols including AODV, DSR, DSDV, OLSR and DYMO. The results of our extensive network simulations are tabulated along with a comprehensive analysis. The effort allows a fair comparison of the capabilities and limitations of different types of mobility patterns and their suitability for contemporary MANET routing protocols. Index Terms—Performance comparison, routing protocols I. INTRODUCTION
Article
The computing progression is towards ubiquitous computing with many interesting happenings and severe challenges. MANET is a collection of intercommunicating mobile hosts forming a spontaneous network without using established network infrastructure and has established promising role in ubiquitous computing. MANET nodes mounted on moving vehicles, wearable devices & as personal medical devices are showing significant applications in the field of ubiquitous healthcare. Performance of MANETs largely depend upon the routing mechanism & nature of mobility. The aim of this paper is to evaluate, the performance of AODV protocol for different mobility models like Random Waypoint, Reference Point Group Mobility and Gauss Markov having varying network load & speed. We have evaluated, the performance metrics like network load, packet delivery fraction, and end-to-end delay with AODV Protocol using NS2 Simulator. Our experimental results suggest that AODV protocol with RPGM mobility model has optimized results for varying network load and speed.
Conference Paper
Mobile ad-hoc networks (MANETS) are formed by a collection of mobile nodes that have the ability to form a communication network without the help of any fixed infrastructure. Because of the nature of these networks, routing protocols play a prominent role in their scalability and overall performance. Due to limited radio transmission range, multiple hops may be required in order to exchange data among the communicating nodes. So, a key requirement of any efficient routing protocol is to find a route between two communicating nodes quickly and with low bandwidth overheads. This study inspects two MANET protocols (i.e. Ad-hoc On-demand Distance Vector routing (AODV) and Dynamic Source Routing (DSR)) and examines their performance based on variation of node density and mobility using mobility models such as Random Way Point (RWP) and Random Way Point with Attractions (RWP-ATTR). The performance is determined on the basis of packet delivery ratio, normalized routing load, throughput and average end-to-end delay with varying node density and mobility. DSR with higher node density shows an extreme degradation in performance. The non-uniform node distribution that occurred for RWP has significant impact on performance results for both protocols.