=Paper=
{{Paper
|id=Vol-2889/PAPER_17
|storemode=property
|title=Performance of AODV, DSR and ZRP for Different Mobility Model in MANET
|pdfUrl=https://ceur-ws.org/Vol-2889/PAPER_17.pdf
|volume=Vol-2889
|authors=Suresh Kumar,Rakesh Sidharth
}}
==Performance of AODV, DSR and ZRP for Different Mobility Model in MANET==
https://ceur-ws.org/Vol-2889/PAPER_17.pdf
Performance of AODV, DSR and ZRP for Different Mobility
Model in MANET
Suresh Kumara and Rakesh Sidhartha
a
ECE Department, University Institute of Engineering and Technology (UIET) MDU, Rohtak, Haryana, India
Abstract
In Mobile ad-hoc Network (MANET), all nodes are self-organized and self-motivated. These
mobile nodes are connected to each other mostly by a wireless link. The schedule of mobility
of these nodes are not fixed and hence not pre-planned. It keeps changing randomly based on
the application. Each Routing protocol (RP) have definite advantages and disadvantages for a
particular chosen performance parameters. In this present paper we have modelled and
evaluated the performance of Routing protocols i.e., Ad-hoc on Demand Distance Vector
(AODV), Dynamic Source Routing (DSR) and Zone Routing Protocol (ZRP) with various
Mobility models (MMs) namely Random Walk Mobility (RWM), Random Way Point
Mobility (RWPM), Group Mobility (GM) and File Base Mobility (FBM). The evaluation
parameters selected are Throughput (mbps), Delay (microsecond) for a configuration of
10and 20 number of nodes by using NETSIM Simulator.
Keywords 1
AODV, DSR, ZRP, Mobility Model, Netsim.
1. Introduction
MANET is basically an independent system of mobile nodes where no fixed infrastructure exists
and network Connectivity keeps changing regularly in random way. In MANET, all mobile units in
the field share information to each other without a central control unit. The Source node and
destination node works as transmitter and receiver respectively. In this layout each nodes is
completely free to move anywhere in the area of responsibility connected wirelessly in field space [1].
No Separate router is available because the function of all the nodes are to act as self-sustained router.
All nodes are capable of finding the routes and maintain the path directory for data transmission from
source to destination nodes. The network structure continuously changing with respect location and
application by executing a particular model of node mobility. In well and highly developed areas,
communication between nodes may get affected due to fading effect. To reduce this effect network
manager uses some common techniques such as (i) Diversity reception (ii) Rake receiver (iii) Space
time coding (iv) Forward error correction etc [2]. A basic systematic layout of MANET Scenario is
shown in figure 1 below.
WCNC-2021: Workshop on Computer Networks & Communications, May 01, 2021, Chennai, India.
EMAIL: skvashist_16@yahoo.com (Suresh Kumar)
ORCID: 0000-0002-3679-1049 (Suresh Kumar)
© 2021 Copyright for this paper by its authors.
Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR Workshop Proceedings (CEUR-WS.org)
161
Figure 1: Basic layout of MANET Scenario
In MANET, Routing protocols (RPs) are divided into two parts: one is unique path RPs and
another is multiple path RPs. AODV, DSR and ZRP are under unique path RPs. AODV and DSR are
Reactive RPs and ZRP is hybrid RP. Reactive RPs can also be denoted as on demand routing
protocol. These Reactive RPs protocols basically work on two things: (i) Route discovery and (ii)
Route maintenance. Whereas the process of route discovery gets executed when routes are required in
the event of disruption of the link between source and destination and has resulted in link failure. This
will lead to commencement of the route search process [3]. These are mainly (i)AODV (ii) Light
Weight Mobile Routing (LMR), (iii) Associativity-Based Routing (ABR) (iv)Temporally Ordered
Routing Algorithm (TORA) (v) DSR, etc. Hybrid RPs are combination of both proactive and reactive
protocols. Few example of these are: (i) ZRP (ii) Zone-Based Hierarchical Link State (ZHLS)
(iii)Distributed Spanning Trees based routing protocol (DST) (iv) Distributed Dynamic Routing
(DDR) (v) Scalable Location Update routing protocol (SLU) etc.[4]. Each RPs have definite
advantages and disadvantages. In the present research the focus has been on AODV, DSR and ZRP
protocols for this present modelled MANET scenario. The goal of this present paper is to find out the
performance of AODV, DSR and ZRP with multiple MMs such as RWM, RWPM, GM and FBM in
term of Throughput and Delay for 10 and 20 nodes. Many places where the MANET have wide
applications are comprises of Defence, Military, War zone, Farming, Medical Robotics automation
etc. [5].
This paper is organized in the order such that the latest research works are given in Section II and
the proposed work is explained in Section III. Section IV presents the results and discussion of the
simulated work. The overall output of this scenario is summarized in Section V.
2. Updated Research Work
In [6] authors analysed the performance of AODV and ZRP RPs at different speed. The simulation
results were calculated for End to End delay, Throughput, Queue length and Drop packets. Author
used Qualnet Simulator for analysis. On the bases of simulation output, AODV performed better than
ZRP.
In [7] the authors have compared the performance of AODV and AOMDV RPs for 40, 80, 120
nodes at maximum speed of 10m/s. The performance metrics evaluated parameters chosen are
throughput (b/s) and average end to end delay (second). Authors found performance of AOMDV is
good in all output parameter as compared to AODV RPs.
In [8] the authors compared AODV, AOMDV and DSDV. Based on the simulation result, authors
analysed that AODV perform better in terms of the (i) throughput, (ii) RO. It has also been seem that
while checking the packet delivery ratio and (iv) packet loss, AOMDV is more reliable. For Delay,
DSDV is more credible than AODV and AOMDV.
In this paper [9] Authors Described three energy model such as Generic, Micaz and Micamotes for
transmitting mode and receiving mode using AODV and Dynamic MANET on Demand (DYMO)
162
RPs. On the basis of simulation outcome authors found that AODV RPs perform better in Micamotes
energy model than other energy model. For throughput and AEED, AODV also performs well.
In [10] the authors evaluated AODV RPs for Throughput, jitter, AEED, Total packet received and
Energy expenditure models. On the basis of Qualnet Simulation Outputs, It has been observed that
jitter is high in Micaz model. In transmitting and receiving mode, Energy consumption of Micamotes
is very less as compared to other energy models.
In [11] the authors used Qualnet simulator version 5.0.2 for simulation. And they compared the
performance for: AODV, DSR and ZRP based on CONSTANT BIT RATE (CBR). The performance
evaluated in terms of first & last Packet transmit (second), Total bytes & packet transmit, Throughput
client & Server (bits/second), First & last Packet received (second), Total received bytes. They found
that these RPs performed good at constant bit rate.
3. Proposed Model
In this proposed network model, simulation is done using NETSIM with Version 9.0. All the
essential work parameters chosen are described in the table 1.
Table 1 : MANET Scenario parameters
Parameters values
Environment area 700*700 meter2
Simulator tool NETSIM 9.0
Routing Protocols AODV , DSR , ZRP
No. of nodes 10 , 20
Mobility types RWM , RWPM , GM , FBM
Maximum mobility speed 15 m/s
Simulation time 120 seconds
Application type CBR
The performance evaluated of these protocols: AODV, DSR and ZRP with multiple MMs namely
RWM, RWPM, GMP and FBM in term of throughput and delay. The performance is calculated for
two different set of 10 and 20 nodes in area of 700*700 meter2. The application applied between
source and destination is CBR type. The maximum speed of nodes is 15 m/s in this network scenario.
In 10 and 20 nodes network scenario. First among the nodes will be acting as source and final node in
the chain as destination node. All nodes are connected to each other by a wireless link. MANET
scenario with 10 nodes is presented in figure. 2 and with 20 nodes is presented in figure. 3
respectively.
Figure 2: MANET scenario with 10 nodes
163
Figure 3: MANET Scenario with 20 nodes.
4. Results & Discussions
4.1 Throughput: Total number of data bit is transferred in specific time duration from Source to
destination is called Throughput and It is measured in mbps. It represents the state of transmitted
information rate in the network [12].Its represented in equation.1
Mathematically
𝑇ℎ𝑟𝑜𝑢𝑔ℎ𝑝𝑢𝑡 = (1)
4.2 Delay: It is the time collected by the data to travel from transmitter to receiver in the network. It
is calculated in microsecond. Various types of delay are included in Delay such as route finding,
propagation and retransmission etc. [12].
Case 1: No. of nodes 10
In this MANET scenario shown in Figure 4, while calculating the throughput verses mobility, it is
observed that for DSR, Throughput are 0.583903, 0.583903, 0.583903 and 0.583903. For AODV,
Throughput are 0.584584, 0.584487, 0.583903 and 0.583903. For ZRP, Throughput are 0.572612,
0.100837, 0.567745 and 0.0567648 for RWM, RWPM, GM and FBM respectively.
Throughput Vs Mobility for 10 nodes
0.7
Throughput
0.6
0.5
0.4
0.3
0.2
0.1
0
RW RWP GM FBM
DSR 0.583903 0.583903 0.583903 0.583903
AODV 0.584584 0.584487 0.583903 0.583903
ZRP 0.572612 0.100837 0.567745 0.567648
Figure 4: Throughput versus mobility model for 10 nodes.
164
Throughput of the AODV and DSR with all mobility model is almost same. But for ZRP,
throughput is less for all mobility model except for group mobility. From overall performance seen
through result is that the AODV and DSR is more reliable than ZRP.
On the basis of simulation outcomes to measure the Delay verses mobility as shown in Figure 5, it
has been found that for DSR, Delay are 12862.28476, 12862.3644, 12860.25648 and 12861.45009.
For AODV, Delay are 16613.86295, 16232.52069, 16254.82639 and 16107.94337. For ZRP, Delay
are 21593.59712, 93018.86365, 18061.32 and 17285.52525 for RWM, RWPM, GM and FBM
respectively.
Delay Vs Mobility for 10 nodes
100000
90000
80000
70000
Delay
60000
50000
40000
30000
20000
10000
0
RW RWP GM FBM
DSR 12862.28476 12862.3644 12860.25648 12861.45009
AODV 16613.86295 16232.52069 16254.82639 16107.94337
ZRP 21593.59712 93018.86365 18061.32 17285.52525
Figure 5: Delay versus mobility model for 10 nodes
From overall result conclusion, DSR is perform better than other two Routing protocols.
Case 2: No. of nodes 20
In this new scenario shown in Figure 6, while calculating the throughput verses mobility, it is
observed that for DSR, throughput are 0.583903 , 0.583903 , 0.583903 and 0.583903, for AODV,
Throughput are 0.585071 , 0.58692 , 0.583903 and 0.592663. For ZRP, Throughput are 0.049543,
0.015379, 0.573585 and 0.056259 for RWM, RWPM, GM and FBM respectively.
165
Throughput Vs Mobility for 20 nodes
0.7
0.6
Throughput
0.5
0.4
0.3
0.2
0.1
0
RW RWP GM FBM
DSR 0.583903 0.583903 0.583903 0.583903
AODV 0.585071 0.58692 0.583903 0.592663
ZRP 0.049543 0.015379 0.573585 0.056259
Figure 6: Throughput versus mobility model for 20 nodes.
Throughput values of AODV and DSR are near about. From overall result conclusion, AODV and
is more reliable than DSR, ZRP. Throughput of the AODV and DSR is approximately same.
From Simulation Result, outcomes to measure the Delay verses mobility as shown in Figure 7, it
has been found that for DSR, Delay are 12924.63905, 13613.1403, 14741.00849 and 13610.61842.
For AODV, Delay are 19868.04275, 20871.71512, 18846.58386 and 24180.38182. For ZRP, Delay
are 90279.63851, 21882.57866, 86130.1058 and 91064.24325 for RWM, RWPM, GM and FBM
respectively.
Delay Vs Mobility for 20 nodes
100000
90000
80000
70000
Delay
60000
50000
40000
30000
20000
10000
0
RW RWP GM FBM
DSR 12924.63905 13613.1403 14741.00849 13610.61842
AODV 19868.04275 20871.71512 18846.58386 24180.38182
ZRP 90279.63851 21882.57866 86130.1058 91064.24325
Figure 7: Delay versus mobility model for 20 nodes.
In 10 nodes, DSR, AODV and ZRP RPs are used to calculate the performance for these MMs. In
the results, throughput is high and delay is low in DSR as compared to both AODV and ZRP RPs.
Throughput of AODV is much closed to DSR but it is less. Therefore DSR is best in both throughput
and Delay metrics with these MMs. In case of 20 nodes, we repeat same process and observed that
throughput is high in AODV, DSR than ZRP RPs. For delay, DSR have low delay compared to other
RPs. So DSR is also best for 20 nodes with these MMs. Therefore we can say that DSR performed
well for these MMs with increase in number of nodes.
166
5. Conclusion
In this present research paper, the Comparison of the performance of AODV, DSR and ZRP
routing protocols with multiple MMs namely RWM, RWPM, GM and FBM in terms of the
throughput (mbps) and delay (microsecond) for a configuration of 10 and 20 nodes has been
presented. These outcome of the performance metrics vary with all MMs. For ZRP, both throughput
and delay are highly unstable with some MMs. But in the case of DSR and AODV, the results are
stable. After deeply analysis of simulation results, we found that the performance of DSR is more
reliable and efficient than both AODV and ZRP RPs with all types of mobility models.
6. References
1. Kumar, S., Kumar, C. D. S. (2015). Study of MANET: Characteristics, Challenges, Application,
Routing Protocol and Security Attacks. INTERNATIONAL, JOURNAL, 2(5).
2. Ekwe, O. A., A. E. Abioye, M. O. Oluwe, and K. C. Okoro. (2014). Effective fading reduction
techniques in wireless communication system. IOSR Journal of Electronics and Communication
Engineering 9, no. 4 (2014): 35-43.
3. Khurana, S.,Kumar, S., & Sharma,D.(2017). Performance evaluation of congestion control in
MANETs using AODV, DSR and ZRP protocols. International Journals of Advanced Research
in Computer Science and Software Engineering, (ISSN: 2277-128X) Vol (2017): 398-403.
4. Abolhasan, M., Tadeusz,W., & Eryk, D.(2004).A review of routing protocols for mobile ad hoc
networks. Ad hoc networks 2, no. 1 (2004): 1-22. https://doi.org/10.1016/S1570-8705(03)00043-
X
5. Nayak, P., Pallavishree,S.(2015). Analysis of random way point and random walk mobility
model for reactive routing protocols for MANET using NetSim simulator. In 2015 3rd
International Conference on Artificial Intelligence, Modelling and Simulation (AIMS), pp. 427-
432. IEEE, 2015. https://doi: 10.1109/AIMS.2015.87.
6. Mehra, M., Mehajabeen F. (2020). AODV and ZRP Analysis for Congestion in MANET.
International Journal of Research in Engineering, Science and Management Volume-3, Issue-5,
May-2020 ISSN (Online): 2581-5792
7. Sharma, D., Kumar,S.(2020). Performance Evaluation of MANETs with Variation in
Transmission Power using Ad-hoc on-demand Multipath Distance Vector Routing Protocol.
In 2020 5th International Conference on Communication and Electronics Systems (ICCES), pp.
363-368. IEEE, 2020. https://doi: 10.1109/ICCES48766.2020.9137954.
8. Setijadi, E., Purnama, K.E. & Pumomo, M. H. (2018). Performance comparative of AODV,
AOMDV and DSDV routing protocols in MANET using NS2. In 2018 International Seminar on
Application for Technology of Information and Communication, pp. 286-289. IEEE, 2018.
https://doi: 10.1109/ISEMANTIC.2018.8549794.
9. Kumar, S., Dhull, K., Sharma, D., Arora, P. & Dahiya, S. (2019). Evaluation of AODV and
DYMO Routing Protocol using Generic, Micaz and Micamotes Energy Conservation Models in
AWSN with Static and Mobile Scenario. Scalable Computing: Practice and Experience 20, no. 4
(2019): 653-662. https://doi.org/10.12694/scpe.v20i4.1584
10. Kumar, S., Dhull, K., Arora, P. & Luhach , A. K. (2019). Performance of Energy Conservation
Models, Generic, Micaz and Micamotes, using AODV Routing Protocol on a Wireless Sensor
Network. Scalable Computing: Practice and Experience 20, no. 4 (2019): 631-639.
https://doi.org/10.12694/scpe.v20i4.1563
11. Kumar, S., Kumar, J. (2012). Comparative performance study of zone routing protocol over
AODV and DSR routing protocols on constant bit rate (CBR). IJCA (097S-8887) Volume 4S-
NoA, (2012).
12. Singh, G., Khurana, R. (2017). Performance Evaluation of DSR, OLSR and ZRP using NetSim
Simulator.International Journal of Advanced Research in Computer Science 8, no. 3 (2017):
346-349.
167