See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/325484919

CSEIT183541 | Impact of Mobility on Routing Protocols in MANET using NS2

Article · June 2018

CITATIONS READS

2 407

3 authors, including:

Meerasharif Sheik Bobba Brao

Godavari Institute of Engineering and Technology Acharya Nagarjuna University
8 PUBLICATIONS 6 CITATIONS 45 PUBLICATIONS 97 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

GPR Objects Hyperbola Region Feature Extraction View project

Comparison between Routing Protocols of MANET using NS2 with and Without queuing parameters View project

All content following this page was uploaded by Meerasharif Sheik on 17 July 2018.

The user has requested enhancement of the downloaded file.

 International Journal of Scientific Research in Computer Science, Engineering and Information Technology
© 2018 IJSRCSEIT | Volume 3 | Issue 5 | ISSN : 2456-3307

Impact of Mobility on Routing Protocols in MANET using NS2

B. Basaveswara Rao#1, SK. Meera Sharief *2, K. GangadharRao#3
#Department of Computer Science and Engineering, Acharya Nagarjuna University, Guntur, Andhra Pradesh,
India
* Department of Computer Science and Engineering, GIET College of Engineering, Rajahmundry, Andhra
Pradesh, India

ABSTRACT

This paper investigates the impact of mobility patterns on different routing protocols in MANET through NS2.
The simulation study is carried out in two schemes and they are classified based on buffer availability at node.
They are (a) a finite buffer is created at each node called Buffer Node (BN Scheme) and (b) buffer facility is not
available at each node called Buffer Less Node (BLN Scheme). In this simulation study Random Way Point
Mobility Model is considered with mobility parameters are speed and pause times are considered for reactive
(AODV and DSR) and proactive (DSDV) routing protocols. The simulation study is performed and to evaluate
three performance metrics for both BN and BLN environments to identify the impact of mobility on routing
protocols in MANET. The performance metrics are Packet Delivery Ratio (PDR), End to End Delay (E-ED) and
Average Routing Load (ARL) are considered. For various configurations selective ten runs were performed for
each result. Finally the results are presented and draw conclusions. Based on these results to get an idea about
the implementation of protocols with respect to mobility of these two schems.
Keywords : MANET, AODV, DSDV, DSR, Mobility Parameters, Queuing Parameters, QoS, CBR, PDR, BN,
BLN, E-ED, ARL.

I. INTRODUCTION nodes can be possible if they are in same radio

transmission range or indirectly by using multi-
Mobile Ad hoc Network is a wireless network of hopping technique. The neighboring nodes act as
mobile routers or hosts formed by the union of intermediate nodes in which source and destination
arbitrary topology which is connected by wireless nodes are not within the same radio transmission
links, and it configures itself by its own. This is range then multi-hop concept will be used. It is very
basically infrastructure less and no centralized difficult to maintain the connectivity between the
controller. The router moves freely from one place to nodes in MANETS because of the intrinsic
another within the area and they organize complexity of routing between any given pair of
themselves randomly. It consists of mobile nodes nodes. Because of dynamic topology of MANET,
which have more mobility without any centralized there are new challenges for routing protocols,
administration and they can join or leave the researchers are comparing and improving existing
network at their own time. Mobile nodes are in the routing protocols using simulations.
state of mobility in the network so that
communication between source and destination These networks can operate autonomously or may be
connected with the help of Internet and then

CSEIT183541 | Received : 05 May 2018 | Accepted : 14 May 2018 | May-June-2018 [ (3)5 : 120-131 ]
120
 B. Basaveswara Rao et al. Int J S Res CSE & IT. 2018 May-June; 3(5) : 120-131

transmitted information with one or more hops achieve this objective the successive contributions
between the nodes. In this decade the growth of have to be done for identifying the impact.
laptops and Wi-Fi utilization has made more  It explores the comparison between the
complexity in routing mechanisms. The mobility is performance of two schemes BN & BLN and also
an essential characteristic of MANETS which amongst the flat topological routing protocols as
necessitates the study of the performance of proactive (DSDV) and reactive (AODV and DSR)
protocols for designing/developing new protocols so routing protocols.
here considered the mobility parameters as Pause  In BN scheme the packet arrival rate and the
time and Speed. service rate at the node kept constant with finite
The performance evaluation of several routing buffer fixed for 30 nodes may be varied pause
protocols in MANET with different traffic loads, time and speed at regular interval.
mobility models, scalability factors, number of nodes  In BLN Scheme number of nodes is kept constant
and different metrics was developed [1, 2, 4, 5, 6 and with a value of 30 and pause time and speed may
7]. It is difficult to analyze or measure the metrics be varied at a regular interval.
and finding the performance metrics of routing  Simulation of AODV, DSR, and DSDV routing
protocols of a MANET without using a queuing protocols performance measures PDR, E-E Delay
model. The limited buffer size in queuing model will and Average Routing Load are studied in 2
be more realistic in calculating Packet Delivery Ratio scenarios i.e. pause time and speed for both the
(PDR), End to End (E-E) Delay and Average Routing schemes i.e. BN and BLN.
Load (ARL). If there is no restriction in buffer size
This paper is organized as: Section 2 discusses the
then the each nodes keep bundles in their buffer and
related research work in flat topological routing
relay them to the nodes they encounter until they
protocols, MANET Models, and Queuing System;
deliver them to the destination, in that case they
Section 3 explains about the Mobility models as
drop them from their buffers. Also when a bundle
MANET Topology, Buffered Node, No Buffer Node
reaches TTL, it is also dropped from the buffer. Now
and Mobility Model; Section 4 discusses about the
that the buffer has a limited size, implement the
Simulation Methodology which describes about
following management policy: when a new bundle
simulation parameters as Queuing Parameters and
arrives at a node and if the buffer does not have room
mobility parameters; In Section 5 in deepness it
for it, it later buffers [8]. It eliminates the oldest
describes about Simulation Result Analysis and
bundle it has in the stock and which has been not
performance of different protocols in the Mobile ad
relayed until that point and replaces the new one. As
hoc Network with and without Queuing Approach
per the dynamic nature of the nodes it is important
and; Section 6 describes about Conclusion and
to study the performance of routing protocols
Future Work of the Research Paper.
through simulation.

BN and BLN schemes were considered for MANET II. Related Work
mobility model and Mobility parameters as Speed
There are different types of performance evaluation
and Pause time .There is fundamental need to
studies that are performed in MANETs for the
simulate and study the creation of real time Mobile
routing protocol with different traffic loads, and
Ad Hoc Networks with, Packet Delivery Ratio, End
their influences of mobility models. The key
to End Delay and Average Routing Load at each node
component of any MANET model is establishing a
should be high, very low and low respectively. To
protected routing with a improved performance

Volume 3, Issue 5, May-June-2018 | http:// ijsrcseit.com

121
 B. Basaveswara Rao et al. Int J S Res CSE & IT. 2018 May-June; 3(5) : 120-131

metrics i.e., high throughput and low end to end A.Lee [4] has developed an adaptive-gossip algorithm
delay. with probability pn for reducing the routing overhead,
over the flooding based routing method in a queuing
Laxmi Shrivastava, [1] has compared the evaluation network model based on ad hoc routing networks for
of routing protocols such as AODV, DSDV and DSR multimedia communications. Pan Li[5] has employed
in mobile adhoc network with different traffic load a practical restricted random mobility model by
as heavy and low , finally concluded that reactive proposing a new multi hop relay scheme for smooth
protocols perform better that proactive protocols that trade-offs between throughput and delay by
is DSR has performed well compared to AODV and controlling the nodes mobility independent of
DSDV in simulation run of heavy traffic load with 40 network size. Saad Talib [6] designed a model for
connections. These protocols are evaluated by means queuing approach of two queue mechanisms (Drop
of number of dropped data packets, average delay, tail and Random Exponential Marking) at each node
packet delivery ratio versus all the routing protocols of network for evaluating the performance of certain
with CBR 20 and 40. MANET parameters. When simulation area is
increased it is observed that low throughput with
S.Mohapatra,[2] has identified the performance REM and Drop tail, REM gives low delay compared
analysis of AODV, DSR, OLSR and DSDV Routing with Drop tail, but in packet loss drop tail is better
protocols using NS2. These protocols were analyzed than REM.
in three Scenarios by using number of nodes vs
Throughput, End to End Delay and Control Over A.H.Zakaria [7] has performed performance analysis
Head. Finally they identified that the DSR protocol is in MANETs using queuing theory by determining its
the best in Packet Delivery Ratio and OLSR is the arrival times, average waiting times and response
optimum at high mobility ratio. time for DSR protocol by varying the input arrival
rates using the queuing system M/G/1. They
Kirti Jain, [3] has compared the influence of mobility observed that if the number of hops increase then
models that is totally different mobility models in the there is an increase in the values of waiting time and
performance evaluation of mobile ad hoc network average response time. Mouna A.[8] have identified
routing protocols such as AODV, DSDV and DSR and two important issues namely encounter and
concluded that these protocols achieve high packet exchange using a limited buffer management
delivery ratio and the least overhead with random constraint for performance of DTMN if there is ill
waypoint model where as low packet delivery ration behaved mobility schemes and its improvement by
and high overhead with city section model. introducing relay nodes for poor performance factor
of DTMN. Rekha[10] has considered AODV, DSDV,
o.o omitola, [9] has performed a performance and DSR for comparison of MANET parameter in
evaluation of AODV, DSR, and TORA routing Packet Delivery Ratio fraction and End to End Delay
protocols in MANETs by varying number of nodes such that increase in density of nodes yields to an
and different metrics with two different scenarios as increase in the Mean E-E Delay and increase in the
conference and event scenarios both of these pause time leads to a decrease in the E-E Delay and
scenarios DSR perform better that AODV and TORA finally the increase in number of nodes will cause an
in terms of throughput, AODV showed a moderate increase in the mean time for loop detection without
result. Interms of delay AODV outperforms DSR and using any queuing model.
TORA. Prachi Jain [11] has developed a new scheme of
buffer management to handle the packet queues in

Volume 3, Issue 5, May-June-2018 | http:// ijsrcseit.com

122
 B. Basaveswara Rao et al. Int J S Res CSE & IT. 2018 May-June; 3(5) : 120-131

MANET and applied the concept of RED (Random protocols AODV, DSR, and DSDV are chosen for the
Early Discard) algorithm on TCP to maintain the study because of the flat routing protocols and they
packets in the buffer and also reaches the destination follow the same configurations .The buffer centric
node with low packet delivery time. schemes are taken from [12] BN and BLN Schemes.
BN and BLN schemes are performed for identifying
B. Basaveswara Rao and SK M Sharief [12] has the impact of mobility on the routing protocols. The
simulated different routing protocols in MANET comparison of routing protocol under several
with and without buffer nodes and their constraints were performed earlier (1, 2, 3, 4, 5, 6, 7,
performance metrics are evaluated and compared in 8) but this paper proposes the buffer schemes under
AODV, DSDV and DSR as reactive and proactive the mobility parameters pause time and speed were
routing. These protocols are performed under the considered for the comparison. In simulation study
influence of buffered node and no buffer node using number of nodes (N), 𝜆 and 𝝻 are fixed for different
different performance metrics as packet delivery scenarios and protocols (AODV, DSDV, and DSR) in
ratio, average routing load and End to End delay by two cases. Where 𝜆 is arrival rate (expected number
varying number of nodes, service rate and arrival rate of arrivals per unit time) and 𝝻 is service
were performed in buffered node scheme where as rate(expected number of transmitting service per
number of nodes was performed in No buffer node unit time). Arrival rate and service rate follows
scheme. exponential distribution. In these parameters one is
varied and the other are kept constant resulting two
III. Preliminaries different cases they are Pause Time increasing
Scenario and Speed increasing Scenario.
MANET Mobility models are different types are
available based on the different strategy and Random Waypoint Model (RWP):
approach. MANET topology is an random model in Mobility model describes the way of mobile nodes
which the nodes are having self-governing mobility travel in the network area. In this, nodes may travel
with a variable network model N, which is a limited in any direction and at any speed based on pause
set of mobile nodes (1,2,3,…N) and the nodes are time, speed and other characteristics. The random
autonomously placed in a throughway scenario with way point mobility model is used as network node
an area of A(a X b) where a is the length and b is the mobility model for the mobile ad hoc network. The
width of the rectangular freeway scenario area. Each mobility model plays a vital role in the evaluation of
node is assumed to have the transmission range Tr. the performance metrics of MANET. In Random way
Let “ld” denote the distance between nodes a and b. mobility model every node has speed and pause time
Nodes a and b are said to be neighbors if theyare in attributes. Initially the nodes are positioned
the same transmission range that is they can directly randomly in the mobile ad hoc network area. The
communicate with each other, that is if ld < Tr . There nodes are moving in the network based on the pause
are N Nodes in the mobile ad hoc network where time and speed of each node until the end of the
node S is the source and node D is destination for the simulation time. Normally mobile nodes travel near
packet transmission. The transmission of data or the center of mobile ad hoc network area or
packets may be transferred or traversed from S to D simulation area. This mobility model happens to be
using intermediate nodes called neighbor node and if widely used one hence chosen.
the Source node S and Destination node D are with
in the same transmission range then there is no Network Simulator:
intermediate node is involved as neighbor nodes. The

Volume 3, Issue 5, May-June-2018 | http:// ijsrcseit.com

123
 B. Basaveswara Rao et al. Int J S Res CSE & IT. 2018 May-June; 3(5) : 120-131

Simulation of computer networks is performed by factor is the average of these 10 output simulation
using a Network Simulator tool. There are different runs. In all the simulation runs each one takes 10
computer network simulator tools are available. The sample points of particular factors and verified three
general networks simulators are ns2, ns3, Omnet++, different protocols i.e. AODV, DSR, and DSDV.
Opnet, Qualnet, Mininet, Onesim, Cooja, Tossim, Hence 150 simulation runs for each case were
Veins, Sumo, Glomosim, Peersim, WSN Simulator, performed to analyze each performance factor for
GNS3, JIST/SWANS ,TETCOS NetSim etc. for these three protocols. The Parameters that are to be
research work in computer networks. It is really a taken in the simulation are divided into three types
difficult task of implementation of large networks as Queuing Parameters, Simulation Parameters and
and establishment of routers in real time and also Mobility Parameters; the same are presented in table
cost effective. It is easy for every researcher to 4.1. Queuing Parameters are arrival rate, service rate
perform the experiments in the network simulator and buffer size of the queue. The simulation
using simulation techniques and programs. NS2 is a parameters are area of the network, number of
network simulator tool version 2 used for network nodes, transmission rate, packet size, simulation
establishment, their implementation, simulation of time, routing protocols. Whereas mobility
real time networks, designing of network protocols, parameters are speed of the node, pause time,
monitoring of node parameters and other parameters mobility model and CBR traffic generation. All the
as required by the user or researcher. The application parameters are common to both BN and BLN
is very easy to implement and perform the Schemes.
experiments in a realistic nature by using NS
Table 4.1: Simulation Parameters
(Network Simulator), the other ancestors of the ns is
ns1, ns2, ns3. NS2 is free software, any user can Queuing Parameter Value
access NS2 for research, development, and work outs Arrival Rate (𝜆 ) 6 packets per second
without any financial overhead in linux and Service Rate ( 𝝻) 15 packets per second
windows platforms. There is no significant difference Buffer Size(K) 10
between the real & simulation results hence no need Simulation Value
of real infrastructure as a cost effective metric. Parameter
Area(A) 1000 m X 500 m.
IV. Simulation Methodology Number of 30
The simulation generally performed based on the Nodes(NN)
preliminaries as discussed in the above and the Transmission Rate 10 packets/sec
objective of simulation is to identify the impact of Packet Size 512 Bytes
mobility and quantify the effects of various factors Simulation Time 200 sec
and their interactions on the overall performance Routing Protocols AODV, DSR, DSDV
before correlating it to the exact image of today’s real Mobility Value
applications. Each run of the network simulator Parameters
accepts input scenario file that describes the exact Speed 5, 10, 15, 20, 25 m/sec
motion of each node and transmission of data using Pause Time 5, 10, 15, 20, 25 sec
BN and BLN schemes with the given pause time and Traffic Generation CBR
speed. The evaluation of the performance at a
particular factor can be measured as the average of 10
random simulation runs for the 10 generated random
scenario patterns. Performance of the considered

Volume 3, Issue 5, May-June-2018 | http:// ijsrcseit.com

124
 B. Basaveswara Rao et al. Int J S Res CSE & IT. 2018 May-June; 3(5) : 120-131

Performance Metrics: PAUSE TIME INCREASING SCENARIO:

The performance metrics are used to quantify the In this scenario three performance parameters were
affect of routing protocol to reach the QoS. To considered as packet delivery ratio, end to end delay
calculate the performance and Quality of Service and Average routing load. Pause time is varied from
issues in MANET, the performance metrics are 5 to 25 with 5 interval time as 5, 10, 15, 20, 25 where
identified as Packet Delivery Ratio, End to End Delay, as the nodes are fixed at 30 number of nodes and
and Average Routing Load [12] constant speed 10. The remaining parameters are
fixed as usual for this case as per parameter table.
Packet Delivery Ratio (PDR) is the ratio of
successfully delivered packets to the destination
Packet Delivery Ratio with out buffer
compared to the total number of packets that are
120
generated or sent from the sender can be given as
100
80

PDR %
PDR=(Received Pkts / Generated Pkts)*100[12] 60 AODV
(1) 40 DSR
20
DSDV
End to End Delay is the difference between the 0
5 10 15 20 25
received time of the packet and the sent time of the
Pause Time
packet.

Fig.1: Pause Time vs Packet Delivery Ratio

ETE Delay [Pkt Id]=Received Time[Pkt Id]-Sent
Time [Pkt Id] [12] (2)
In Fig.1 if there is increase in the pause time , all the
protocols performed a constant performance at their
Average Routing Load defined as the total number of
level but in the comparison at any given pause time
routing packets (Control packet sent + Forward)
AODV performed nearly 100% packet delivery ratio
transmitted per data packet delivered at the
as compared with other protocols. Whereas DSR
destination.
protocol performed well as compared with DSDV
ARL=((Control Pkts Sent+Control Pkts
protocol.
Forward)/Data Pkts Received)*100 [12] (3)

V. Results and Discussions End to End Delay with out Buffer

180
In this study the performance of the MANET was 160
140
End to End Delay

evaluated by applying two types of schemes and the 120

results are presented below by scheme wise. 100
AODV
80
60 DSR
BLN SCHEME: 40 DSDV
In this scheme buffer was not available at each node 20
and the pause time and speed are varied from 5 to 25 0
5 10 15 20 25
with an interval of 5.The scheme is extended by two
Pause Time
scenarios namely pause time increasing scenario and
speed increasing scenario. Fig.2: Pause Time vs End to End Delay

Volume 3, Issue 5, May-June-2018 | http:// ijsrcseit.com

125
 B. Basaveswara Rao et al. Int J S Res CSE & IT. 2018 May-June; 3(5) : 120-131

In Fig.2 if there is increase in the pause time, from 5

Packet Delivery Ratio
to 20 DSDV protocol performed low delay as
120
compared with other protocols. Whereas DSR
100
protocol very high at 20 and low at 25 compared
with other protocols. 80

PDR %
60 AODV

Average Routing Load without Buffer 40 DSR

3.5 20 DSDV
Average Routing Load

3 0
2.5 5 10 15 20 25
2 Speed
DSR
1.5
1 AODV
Fig.4: Speed vs Packet Delivery Ratio
0.5 DSDV
0
5 10 15 20 25 In Fig.4 if there is increase in the speed, AODV
Pause Time performed nearly 100% packet delivery ratio as
compared with other protocols. In this DSR protocol
Fig.3: Pause Time vs Average Routing Load performed well as compared with DSDV protocol.

In Fig.3 if there is increase in the pause time, all the

End to End Delay
protocols performed a different performance at their
150
level but in the comparison at any given pause times,
End to End Delay

AODV protocol performed low routing load as

100
compared with other protocols. AODV
50 DSR
SPEED INCREASING SCENARIO
DSDV
0
In this scenario three performance parameters were 5 10 15 20 25
considered as packet delivery ratio, end to end delay Speed

and Average routing load. Speed is varied from 5 to

Fig.5: Speed vs End to End Delay
25 with 5 intervals time as 5, 10, 15, 20, 25 where as
the nodes are fixed at 30 number of nodes and
In Fig.5 if there is increase in the speed, from 5 to 25
constant pause time 10. The remaining parameters
DSDV protocol performed low delay as compared
are fixed as usual for this case as per parameter table.
with other protocols. Whereas DSR protocol very
high at 20 compared with other protocols.

Volume 3, Issue 5, May-June-2018 | http:// ijsrcseit.com

126
 B. Basaveswara Rao et al. Int J S Res CSE & IT. 2018 May-June; 3(5) : 120-131

Average Routing Load Packet Delivery Ratio

7 120

Packet Delivery Ratio

Average Routing Load

6 100
5 80
4 60 AODV
AODV
3 40 DSR
DSR
2 20 DSDV
1 DSDV 0
0 5 10 15 20 25
5 10 15 20 25 Pause TIme
Speed
Fig.7: Pause Time vs Packet Delivery Ratio
Fig.6: Speed vs Average Routing Load
In Fig.7 if there is increase in the pause time, in the
In Fig.6 if there is increase in the speed, from 5 to 15 comparisons at any given pause time DSR performed
DSR protocol performed low routing load as nearly 100% packet delivery ratio as compared with
compared with other protocols. From 20 to 25 DSDV other protocols. DSR maintains the constant packet
performed low routing load as compared with other delivery ration during the buffer scheme and also
protocols. packet delivery ratio is also high.

BN SCHEME: (QUEUING APPROACH)

In this scheme a buffer was created at each node and End to End Delay
the pause time and speed are varied from 5 to 25 250
End to End Delay

with an interval of 5.The scheme is extended by two 200

cases namely pause time increasing scenario and 150 AODV
speed increasing scenario. 100
DSR
50
DSDV
PAUSE TIME INCREASING SCENARIO 0
5 10 15 20 25
Pause Time
In this scenario three performance parameters were
considered as packet delivery ratio, end to end delay Fig.8: Pause Time vs End to End Delay
and Average routing load as in the above case. Pause
time is varied from 5 to 25 with 5 interval time as 5, In Fig.8 if there is increase in the pause time, from 5
10, 15, 20, 25 where as the nodes are fixed at 30 to 25 DSDV protocol performed low delay as
number of nodes and constant speed 10. The compared with other protocols. Whereas DSR
remaining parameters are fixed as usual for this case protocol very high at 20 compared with other
as per parameter table. protocols. DSDV is considered to be best for this
comparison.

Volume 3, Issue 5, May-June-2018 | http:// ijsrcseit.com

127
 B. Basaveswara Rao et al. Int J S Res CSE & IT. 2018 May-June; 3(5) : 120-131

In Fig.10 if there is increase in the speed, DSR

Average Routing Load
performed nearly 100% packet delivery ratio as
3.5
compared with other protocols at any given speed
3
Average Routing Load

time. In this comparison DSR is best and optimal.

2.5
2
AODV
1.5 End to End Delay
DSR
1 100
DSDV

End to End Delay

0.5 80
0 60
5 10 15 20 25 AODV
40
Pause Time DSR
20
DSDV
Fig.9: Pause Time vs Average Routing Load 0
5 10 15 20 25
Speed
In Fig.9 if there is increase in the pause time, all the
protocols performed a different performance at their Fig.11: Speed vs End to End Delay
level but in the comparison at any given pause times,
DSR protocol performed low routing load as In Fig.11 if there is increase in the speed, from 5 to
compared with other protocols except at 20 pause 25 DSDV protocol performed low delay as compared
time AODV performed low routing load. with other protocols. Whereas DSR protocol very
high at 20 and 25 compared with other protocols. In
this comparison DSDV is optimal.
SPEED INCREASING SCENARIO
In this scenario three performance parameters were
considered as packet delivery ratio, end to end delay Average Routing Load
and Average routing load. Speed is varied from 5 to 6
Average Routing Load

5
25 with 5 intervals time as 5, 10, 15, 20, 25 where as
4
the nodes are fixed at 30 number of nodes and
3 AODV
constant pause time 10. The remaining parameters 2 DSR
are fixed as usual for this case as per parameter table 1
DSDV
0
5 10 15 20 25
Packet Delivery Ratio Speed
120
100 Fig.12: Speed vs Average Routing Load
80
PDR %

60 AODV In Fig.12 if there is increase in the speed, from 5 to

40 DSR 15 DSR protocol performed low routing load as
20
DSDV compared with other protocols. From 20 to 25 DSDV
0
performed low routing load as compared with other
5 10 15 20 25
Speed protocols.
The suggested protocols with two schemes for three
Fig.10: Speed vs Packet Delivery Ratio performance metrics are given in the following table
as per the above results.

Volume 3, Issue 5, May-June-2018 | http:// ijsrcseit.com

128
 B. Basaveswara Rao et al. Int J S Res CSE & IT. 2018 May-June; 3(5) : 120-131

Table 5.1: Matrix for Selection Routing Protocols In  Pause Time Increasing Scenario:
Terms Of Mobility Parameters in BN and BLS  In Packet delivery ratio if increasing the pause
schemes time in no buffer scheme AODV performed
SPEED PAUSE TIME excellent performance throughout the scenario
BUFFER BUFFER BUFFER BUFFER where as DSR protocol performed the same
LESS NODE NODE LESS NODE NODE tremendous performance throughout the no
5 1 1 2 2 5 1 1 2 2 5 1 1 2 2 5 1 1 2 2 buffer scheme at any give pause time.
0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5
 In Average routing load if increasing the pause
P D D D A A D D D D D A A A A A D D D D D
time of each node in no buffer scheme DSR
D S S S O O S S S S S O O O O O S S S S S
protocol performed the best compared to other
R R R R D D R R R R R D D D D D R R R R R
protocols where as in buffer scheme also DSR
V V V V V V V
A D D D D D D D D D D D D D D D D D D A D protocol performed well compared to other
R S S S S S S S S S S S S S S S S S S O S protocols except at pause time 25 AODV did well
L R R R R D R R R D D R R R R R R R R D R compared to the remaining.
V V V V  In End to End Delay if increasing the pause time
E D D D D D D D D D D D D D D D D D D D D from 5 to 15 at each node in no buffer scheme
E S S S S S S S S S S S S S S S S S S S S and buffer scheme DSDV protocol performed a
D D D D D D D D D D D D D D D R D D D D D constant low delay compared to the remaining
V V V V V V V V V V V V V V V V V V V
protocols.

As per the three performance metrics are calculated

through simulation study with two schemes for VI. CONCLUSION
random way point mobility model with speed
increasing scenario and pause time increasing In the modern years, Mobile Ad Hoc Networks
scenario, the following observations are drawn from (MANETs) are sky-scraping in research and topology
the above table as well as graphs. of MANETs usually changes with time. The prime
Speed Increasing Scenario: idea about alter in routing protocols or designing of
 In Packet delivery ratio if increasing the speed new protocols has been done in recent years. Because
from 5 to 15 in no buffer scheme DSR protocol of the dynamic nature of MANETs and involved
performs well and then from 20 to 25 AODV complexity to arrive at performance metrics without
performed well compared to other protocols assumptions, there is always a need to explore the
where as in buffer scheme DSR protocol performance of routing protocols through simulation.
performed 100% packet delivery at any speed. The simulation study is carried on two scenarios with
 In Average routing Load if increasing the speed two schemes of three performance metrics. The
of the node from 5 to 20 DSR protocol performed observations from simulation study are as follows
well and at speed 25 DSDV performed low o The increase in the pause time of nodes in no
routing load where as in buffer scheme DSR buffer node scheme leads to an increase in the
performed low from speed 5 to 15 and from Packet Delivery Ratio, decrease in the End to
speed 20 to 25 DSDV protocol performed low End Delay and decrease in Average Routing Load
routing load with increasing speed as compared in AODV routing protocol as per the simulation
to other two protocols. results and graphs. Increase in the pause time
 In End to End Delay at any given speed DSDV would obviously mean reduced congestion which
performed meticulous performance throughout would have a positive impact on all the
the no buffer and buffer schemes. dependant parameters.

Volume 3, Issue 5, May-June-2018 | http:// ijsrcseit.com

129
 B. Basaveswara Rao et al. Int J S Res CSE & IT. 2018 May-June; 3(5) : 120-131

o The scenario of increase in the speed of the nodes [2]. S Mohapatra, P.Kanungo, Performance Analysis of
in no buffer node scheme will give 100% of AODV, DSR, OLSR and DSDV Routing Protocols
Packet Delivery Ratio in DSR &AODV protocols. using NS2 Simulator, ICCTSD 2011, 1877-77058 ©
2011 Published by Elsevier Ltd.
Where as in the case of DSDV protocol the PDR
[3]. Kirti Jain, Rajiv Ranjan Tewari, Influence of
remains decreasing from 80% to 40% . The End
Mobility Models in Performance Evaluation of
to End Delay and Average Routing Load
MANET Routing Protocols, Special Issue of
increases gradually for all the routing protocols if International Journal Computer Applications (0975-
packet delivery ratio is only the factor to 8887).
consider in the network then this is best [4]. Ahyoung Lee & Iikyeun Ra, A Queuing Network
technique to apply in the MANETs. Model Based on Ad Hoc Routing Networks for
o The increase in the pause time of nodes in buffer Multimedia Communications,
node scheme leads to the DSR protocol 100% Appl.Math.Inf.Sci.6No.1S pp.271S-283S.
Packet Delivery Ratio, decrease in the End to [5]. Pan Li, Yuguang Fang, Jie Li , & Xiaoxia Huang,
Smooth Trade-offs between Throughput and Delay
End Delay and decrease in Average Routing Load
in Mobile Ad Hoc Networks. IEEE Transactions on
as per the simulation results and from the
Mobile Computing, Vol. 11, No. 3, March 2012.
obtained graphs. Increase in the pause time
[6]. Dr Saad Talib Hasson and Enass Fadil, Queuing
would obviously mean reduced congestion in the Approach to Model the MANETs Performance,
buffer to enhance the positive impact on all the British Journals ISSN 2047-3745.
dependant parameters. [7]. Aznida Hayati Zakaria, Md.Yazid Mohd Saman,
o The scenario of increase in the speed of the nodes Ahmad Shukri M Noor, Ragb O.M. Saleh,
in buffer node scheme will give 100% of Packet Performance Analysis of Mobile Ad Hoc Networks
Delivery Ratio in DSR protocol .Where as in the using Queuing Theory, DOI:10.1007/978-981-4585-
case of DSDV protocol the PDR remains 18-7_62, © Springer Science +Business Media
Singapore 2014.
decreasing from 80% to 45% and in AODV it
[8]. Mouna Abdelmoumen, Mounir Frikha, Tijani
remains constant at 80% . The End to End Delay
Chahed, Performance of Delay Tolerant Mobile
and Average Routing Load increases linearly for
Networks and its improvement using mobile relay
all the routing protocols. nodes under buffer constraint, 978-1-4673-7468-
o These simulation results proves that the impact 2/15© 2015 IEEE.
of mobility on routing protocols in MANET, the [9]. Oo.omitola, Performance Evaluation of Routing
DSR is performed the best results in packet Protocols in MANETs using Varying Number of
delivery ratio, end to end delay and also in Nodes and Different Metrics, African Journal of
average routing load in both schemes. The future Computing & ICT, Vol 8. No.2 Kune, 2015-IEEE
extension may be performed using different [10]. Rekha, Gurjeet Kaur, Mahipal Singh, Avtar Singh,
Comparative Analysis of Different Mobile Ad Hoc
mobility models and other performance metrics.
Network Protocols.,Research Article Volumen 6
Issue No.7, ISSN 2321 3361© 2016 IJESC.
VII. REFERENCES [11]. Prachi Jain & Vijay Prakash, A Buffer Management
Scheme to Control Packet Loss in MANET., 2016
[1]. Prof. Laxmi Shrivastava, Sarita S. Bhaduaria, Symposium on Colossal Data Analysis and
Performance Evaluation of Routing Protocols in Networking (CDAN),978-1-5090-0669-
MANET with different traffic loads, International 4/16/©2016IEEE.
conference on Communication Systems and [12]. B BasaveswaraRao, SK. Meera Sharief, K.
Network Technologies, 978-0-7695-4437-3/11© GangadharRao, “Comparison between Routing
2011 IEEE DOI 10.1109/CSNT.2011.10. Protocols of MANET using NS2 with and without
queuing parameters”, International Journal of

Volume 3, Issue 5, May-June-2018 | http:// ijsrcseit.com

130
 B. Basaveswara Rao et al. Int J S Res CSE & IT. 2018 May-June; 3(5) : 120-131

Scientific Research and Review Volume 7, Issue 2, Hardware Research in Engineering,IJSHRE Vol 1,
2018 2279-543X. issue 1 September 2013.
[13]. Tonguz O and Ferrari G., “Ad hoc Wireless [25]. Wu Wang, Bin Yang, Osamu T, X Jiange, Shikai
Networks – A communication-Theoretic Shen, On the Packet Delivery Delay Study for
Perspective, Wiley and Sons”, 2009. Three- Dimensional Mobile Ad Hoc Networks,
[14]. FR.B. Cruz, T. Van Woensel, J.MacGregor Smith, Journal Ad Hoc Networks, Vol 69, Issue C, Feb
“Buffer and throughput trade-offs in M/G/1/K 2018, 10.10.16/j.adhoc.2017.10.004.
queuing networks: A bi-criteria approach”, [26]. Yefa Mai, Yuxia Bai, Nan Wang, Performance
International Journal Production economics, 2010. Comparison and Evaluation of the Routing
PP 224-234. Protocols for MANETs Using NS3, Journal of
[15]. CJinshong Hwang, Ashwani Kush, Ruchika, Electrical Engineering 5 (2017) 187-195,
“Performance Evaluation of Manet Using Quality of DOI:10.17265/2328-2223/2017.04.003.
Service Metrics”, Fifth International conference on [27]. Davinder Singh Sandhu and Sukesha Sharma,
Innovative Computing Technology (INTECH 2015) Performance Evaluation of DSDV, DSR, OLSR,
978-1-4673-7551-1/15 © 2015 IEEE. TORA, Routing Protocols- A Review, AIM/CCPE
[16]. Martin Appiah, “PERFORMANCE COMPARISON 2012, CCIS 296, pp.502-507, 2013.© Springer-
OF MOBILITY MODELS IN MOBILE ADHOC Verlag Berlin Heidelberg 2013.
NETWORKS (MANET)”, 978-1-5386-3831-6/17©
2017 IEEE.
[17]. Nsnam web pages: https://www.isi.edu/nsnam/ns/
[18]. Tutorial for Simulation-based Performance Analysis
of MANET Routing Protocols in ns-2 By Karthik
sadasivam
[19]. Teerawat Issariyakul, Ekram Hossain “Introduction
to network simulator NS2”, 10.1007/978-0-387-
71760-9© Springer Science + Business Media, LLC.
[20]. Hui Xu, Xianren Wu, “A Unified Analysis of
Routing Protocols in MANETs”, IEEE Transactions
on Communications, Vol.58, N0.3, March 2010
[21]. Kapang Lego, Pranav Kumar Singh,“Comparative
Study of Adhoc Routing Protocol AODV, DSR and
DSDV in Mobile Adhoc NETwork” Indian Journal
of Computer Science and Engineering Vol.1 No.4
364-371.
[22]. Radhika Patel, Bimal Patel, Sandip Patel, Amit
Parmar, “Comparative Performance Analysis of
Reactive routing protocols, TORA and AODV: A
simulation based evaluation”, Conference Paper
2017.
[23]. Bouchra M., Hicham T., EL Habib B., Md. Talea,
Performance Analysis of Routing Protocols in
Vehicular Ad Hoc Network, Springer Singapore,
DOI:10.1007/978-981-10-1627-1_3.
[24]. Mandeep Singh, Balwinder Singh, Performance
Evaluation of MANET Routing Protocols with
Scalability for HTTP Traffic Delay Using OPNET
Modeler, International Journal of software &

Volume 3, Issue 5, May-June-2018 | http:// ijsrcseit.com

131

View publication stats