Performance of AODV Routing Protocol with increasing the MANET Nodes and it’s effects on QoS of Mobile Ad hoc Networks

Abstract— A mobile ad hoc network (MANET) consists of mobile wireless nodes. The communication between these mobile nodes is carried out without any centralized control. The ease of deployment and the infrastructure less nature of Mobile Ad hoc Networks (MANETs) make them highly desirable for the present day multi media communications. Traditional routing protocols may not suffice for real time communications it depends upon the conditions and our requirements. Though there has been considerable research in this area. In this paper, we are analyzing the performance of reactive routing protocol via increasing number of nodes and observing its effect on Quality of Service (QoS) of Mobile Adhoc Network. As we know routing protocols make an important role for improving QoS in Mobile Adhoc Network. The QoS depends upon several parameters like end-end delay, throughput, date drop and network load. The reactive routing protocol which we are considering is AODV for this scenario with MCHG [1] [2]. Here we are observing performance of Routing Protocol via enhancing the network size on the basis of following parameters: delay, throughput, traffic sent, traffic received, data dropped and network load. Network simulation tool used in simulation is OPNET Modeler (Ver. 14.0). Finally, this paper conducts simulation experiments in the conditions where we can improve QoS of MANET Network performance.

Keywords- Multiple Cluster Head Gateway, QoS, Throughput, AODV, End-End Delay, OPNET.

I. INTRODUCTION MANET stands for Mobile Ad hoc Network. It is a

decentralized autonomous wireless system which consists of free nodes. MANET sometimes called mobile mesh network, is a self configurable wireless network. Nodes are grouped into distinct or overlapping clusters. Clustering provides a Hierarchical MANET system which assists in making the routing scalable. Some of the nodes are elected to be part of the backbone for the MANET system [3] [4] [11] [13-15]. These nodes are called Cluster Head Gateway. Here we are using Multiple cluster Head Gateway (MCHG) with suitable routing protocol in order to enhance the Quality of Service of MANET. AODV we are considering as routing protocol and taking its advantages in our approach to make existing network more reliable and efficient. MANET requires efficient routing algorithm in order to reduce the amount of signaling introduced due to maintaining valid routes[5][6], and therefore enhance the overall performance of the

MANET system. As we can say a Cluster Head Gateway election in order to distribute the load among multiple hosts in the cluster [2].

In this paper we are analyzing the performance of reactive routing protocol via enhancing number of nodes and observe how it effects to QoS of existing mobile Adhoc network. Here Mobile ad-hoc network are dividing into clusters. Each cluster has Manet node with CHG. From one cluster to another cluster or within the cluster we applied reactive routing protocols specifically AODV to evaluate AODV protocol behavior and performance and check what kind of effect made by particular protocol on QoS.

The rest of the paper is organized as follows: Section II presents the related work. Section III describes our proposed working model. Section IV presents the simulation experiment setup and gives the performance evaluation of our proposed strategy. Section V concludes the paper.

II. RELATED WORK Several mechanisms of Cluster Head election exist with

an objective to provide stable and efficient routing in the MANET system [5] [9] [14-15]. Some mechanism assigns the Cluster Head based on the node id as in the Linked Cluster Algorithm (LCA) which selects as the Cluster Head the node with the highest ID [4]. Other mechanisms favor allowing some type of fair share of Cluster Head responsibility by changing the Cluster Head based on an assigned ID to the Cluster Head [3]. A node with a high mobility rate is higher than the duration of Cluster Head rotation may not get the chance to become a Cluster Head. Other Cluster Head election mechanisms consider relative node mobility to ensure routing path availability [4] [10] [12]. However, causing an added signaling overload and causing the elected Cluster Head to pay the higher resource utilization penalty. We can conclude from the existing research that several tradeoffs exist for the elected Cluster Head and the other cluster nodes. • Firstly, the Cluster Head has to bear higher resource

utilization such as power, which may deplete its battery sooner than other nodes in the cluster.

• Secondly, despite fair share responsibility of Cluster Head role, it is possible that heavy burst of traffic takes place causing some Cluster Heads to use maximum resources.

Ashish Bagwari (IEEE Member, MIR Labs

Member) Electronics and Communication

Dehradun, India ashishbagwari@gmail.com

Raman Jee Electronics and Communication

Graphic Era University Dehradun, India

ramanj35@gmail.com

Pankaj Joshi Electronics and Communication

Graphic Era University Dehradun, India

pankajjoshiji@rediffmail.com

Sourabh Bisht (Research Scholar)

Electronics and Communication Graphic Era University

Dehradun, India Sourabh_bisht2002@yahoo.co

2012 International Conference on Communication Systems and Network Technologies

978-0-7695-4692-6/12 $26.00 © 2012 IEEE

DOI 10.1109/CSNT.2012.76

316

2012 International Conference on Communication Systems and Network Technologies

978-0-7695-4692-6/12 $26.00 © 2012 IEEE

DOI 10.1109/CSNT.2012.76

321

2012 International Conference on Communication Systems and Network Technologies

978-0-7695-4692-6/12 $26.00 © 2012 IEEE

DOI 10.1109/CSNT.2012.76

320

• Thirdly, the fair share or load balancing technique [2], might result in a cluster Head that will not provide the optimal path for routing, or yet a link breakage.

There is no one common Cluster Head election mechanism that is best for MANET systems, without some tradeoffs.

The Zone Routing Protocol, ZRP, in [8], provides a hybrid approach which produces added routing control messages in the network due to keeping up to date routes [8]. ZRP divides the network into overlapping zones, while clustering can have distinct, non overlapping clusters. In addition, the authors in [7-8] claim this hybrid approach is suited for large networks, enhances the system throughput, but adds more complexity. As discussed above, the main focus of the previous work focuses on an election of one Cluster Head for a cluster. It mainly can make the central Cluster Head a bottleneck.

III. PROPOSED WORKING MODEL The base of our Paper is trying to improve the QoS of

MANET. For this we are focusing over routing protocol that can bear the access of nodes quantity without degrading the existing system performance. For this we consider AODV as a routing protocol and analyzing its performance along with enhancing number of mobile nodes. In the given figure 1, we have MCHG with nodes having AODV protocol for data communication. Using MCHG, we are distributing the load of the Cluster Head Gateway amongst multiple Cluster Head Gateways in the same cluster and also provide link between nodes. Any of the prior work can be used to select the Cluster Head Gateways for a cluster. In case of one Cluster Head Gateway per cluster, a link breakage caused by the failure of the Cluster Head Gateway isolates all cluster nodes from communicating to/from outside the cluster. However, this reduces the link breakage to be only in the direction towards a path where the failed Cluster Head Gateway forwards the data. Therefore, the reliability of routing in the MANET system is increased.

Figure 1. Showing the MCHG with Nodes using AODV routing protocol As shown in figure 1, Cluster 1 is the one which has a

cluster with multiple Cluster Heads Gateways. The remaining clusters 2, 3 and 4 operated with one Cluster Head

Gateway. AODV is using as a routing protocol in whole existing network. In figure routing from cluster 1 to cluster 2 uses a Cluster Head Gateway N1A which is different from cluster 1 to cluster 3 & cluster 4 use N1B & N1C respectively. Therefore, since there are 3 neighboring clusters to cluster 1, the system allowed for the use of 3 Cluster Head Gateways, one for routing to/from each neighboring cluster.

In our topology, to make better communication, we are using MCHG for different networks and Nodes within network. For this i.e. communication between one cluster to another or within the same cluster we are focusing over appropriate routing protocols which will be helpful to find out the desire route for data communication. To make such kind of communication we are considering AODV as a routing protocol. Here we are analyzing the behavior of AODV routing protocols via varying the number of nodes and check how it effecting to QoS based on the throughput, network load, end-end delay and data drop they all are the parameters to check QoS of Mobile Adhoc Network.

A. Routing Protocols of Ad-hoc Network [16] Routing means to choose a path. Routing in MANET

means to choose a right and suitable path from source to destination. They are of three types:-Reactive Routing Protocols, Proactive Routing Protocols and Hybrid Routing Protocols.

Reactive Routing Protocols are called when they are needed and the routes are built. These routes can be acquired by sending route requests through the network. Disadvantage of this algorithm is that it offers high latency in searching a network. Example AODV and DSR.

In Proactive Routing Protocols the routing information about all the nodes is build and maintained by the proactive protocols. The proactive routing protocols are independent of whether or not the route is needed [17]. Because of the control messages proactive routing protocols are not bandwidth efficient. Example DSDV.

Hybrid Routing, commonly referred to as balanced hybrid routing, is a combination of distance-vector routing, which works by sharing its knowledge of the entire network with its neighbors and link-state routing which works by having the routers tell every router on the network about its closest neighbors. They are combination of both proactive and reactive protocols. � AODV (Ad hoc On-demand Distance Vector)

AODV-node informs its neighbours about its own existence by constantly sending “hello messages” at a defined interval. This enables all nodes to know the status about their neighbours, i.e. if they gone down or moved out of reach. To resolve a route to another node in the network AODV [18] floods its neighbours with a route request (RREQ). A RREQ contain the senders’ address, the address of the sought node and the last sequence number received from that node if there exist one. The receiving node checks if it has a route to the specified node. If a route exists and the sequence-number for this is higher than the supplied a new

317322321

route is found. The node replies to the requesting by sending a route reply (RREP). If on the other hand a route does not exist the receiving node sends a RREQ itself to try to find a route for the requesting node.

IV. SIMULATION SETUP AND RESULTS DISCUSSION

A. Simulation Setup To simulate our Cluster Head Gateway Network, we

used Opnet 14.0 v. The simulation parameters and their values are given in Table 1.

TABLE I. SIMULATION PARAMETERS

PARAMETER VALUE

Total Number of Nodes 45 and 60

Number of MANET Nodes 30 and 40

Number of Packet (Traffic) Sources

4 and 8

Number of Cluster Head Gateway (CHG)

15 and 20

Size of Area 1000*1000 (m.)

Transmission Range 250 (m.)

Traffic Type Constant Bit Rate (CBR)

Protocol AODV

Standard Ad hoc Speed 10 m/s

Address IP v4

Simulation Time 300 (sec)

Wireless Channel Bandwidth 1000 (KHz)

Node Movement Model Default Random Way Point

Data rate 1 (Mbps)

Buffer Size 256000 (Bits)

Frame Size 4 (m.sec)

Datagram switching rate 500,000 (packets/sec)

• Performance Metrics The following metrics are used in varying scenarios to evaluate the different protocols: 1) Throughput – Throughput or network throughput is the

average rate of successful message delivery over a communication channel.

2) End-End Delay - The packet end-to-end delay is the time

of generation of a packet by the source up to the destination reception. So this is the time that a packet takes to go across the network.

3) Packet delivery rate - The total number of data packets received divided by the total number of data packets originated.

4) Data Dropped – This is the difference between total

number of packet transmitted by transmitter and total number of packet received by receiver at receiver end.

5) Network load - The total number of routing messages

transmitted divided by the total number of data packets received.

B. Result Discusiion The Performance of the proposed CHG N/w in Figure 2

to 7 shows with respect to the end-end delay, throughput, traffic sent, traffic received, data dropped and network load respectively. The performance is first evaluated by CHG Internet connectivity.

Figure 2. End-End Delay Vs Time Duration Figure 2 showing End-End Delay Vs time Duration

graph. As we know the value of delay should be as small as possible. With the help of given figure we can say when Number of nodes are less (45 nodes) the delay is also less i.e. 2.0 sec (maximum) at 5 minute time period. At another end when nodes are increasing (having 60 nodes) delay is increasing i.e. around 2.75 sec which is slightly more as compare to previous one but acceptable for the same time period.

Figure 3 showing the Throughput of the whole network with respect to time Period. The value of throughput should be high and from figure we may conclude when number of nodes are 60, we are getting higher throughput value i.e. between 8,000,000 and 11,000,000 (bits/sec) up to 5 min. time periods and 13,000,000 bits/sec maximum at 2.2 min. High throughput is also provide better connectivity between the nodes. While numbers of nodes are fewer (45 nodes) throughputs is less.

318323322

Figure 3. Throughput Vs Time Duration Figure 4 shows Traffic Sent Vs time Duration graph. In

the given figure AODV with large number of nodes (60 nodes) are sending more data as compare to less number of nodes (45 nodes).

For 60 nodes value is between 540 (packets/sec) and 45 nodes value is between 350 (packets/sec).

Figure 4. Traffic Sent Vs Time Duration Figure 5 shows Traffic Received Vs time Duration graph.

In the given figure network is receiving more data when number of nodes are more (60 nodes) as compare to less number of nodes (45 nodes).

For 60 nodes the value is varying between 350 and 400 (packets/sec) and for 45 nodes value is between 100 and 150 (packets/sec).

Figure 5. Traffic Received Vs Time Duration Figure 6 shows Data Dropped Vs time Duration graph.

Analyzing traffic sent and traffic received we can easily calculate the packet drop that should be as minimum as possible and it should be as minimum as possible.

Here using Figure 6 we can say nodes-60 having less data dropped as compare to nodes-45.

Figure 6. Data Dropped Vs Time Duration Figure 7 shows Network Load Vs time Duration graph.

Here Network Load is less with more number of nodes (60- nodes) and more with less number of nodes (45-nodes) with AODV which is responsible for network congestion and it is value should be as minimum as possible. Here AODV with more number of nodes having lesser network load than others.

319324323

Figure 7. Network Load Vs Time Duration

V. CONCLUSION In this paper, we have analyzed Performance of AODV

Routing Protocol with increasing the MANET Nodes and its effects on QoS of Mobile Ad hoc Networks. One of the distinguishing characteristics of our strategy is that, the Mobile Adhoc network is provided a better QoS with appropriate routing protocol with increasing the nodes. In addition, routing reliability is increased since a failure of one Cluster Head Gateway (CHG) does not break all routing to outside the cluster due to use of Multiple CHG. Analyzing figures, in case of End-End Delay, Throughput, Packet drops & Network load for the protocols AODV in our simulation. Finally, simulation results confirm that AODV giving better performs under such types of circumstances, providing better QoS based on good throughput and acceptable End-End Delay, less data drops. One of the notable features of this AODV protocol strategy is that, it reduces our network load which can be responsible for congestion at the time of communication. Therefore it can be used to extend the network coverage. In future the CHG approach can be evaluated under different mobility scenarios and the performance can be measured with other large N/w’s.

ACKNOWLEDGMENT

The authors wish to thank their parents for Supporting and Motivating for this work. And a special thanks to Mr. Pradeep Kumar Bagwari and Mrs. Saraswati Bagwari because without their support this was not possible.

REFERENCES

[1] Ashish Bagwari (IEEE Member), Sourabh Bisht, “Cluster Head Gateway approach using in Integrated Mobile Ad hoc Network”, RAICS-2011: \"IEEE International Conference on Recent Advances in Intelligent Computational Systems\", pp 652-655, September 2011, ISBN: 978-1-4244-9477-4.

[2] Ahmad Anzar, Joshi Pankaj, Rathi Vikas, Bagwari Ashish (IEEE Member), “A Cluster Head Gateway Approach for Deciding the

Cluster head in Mobile Adhoc Network” Singapore Conference December 2011 ICCNS 2010: \"International Conference on Computer Networks and Security\".

[3] A. Amis, and R. Prakash, March 2000. Load- Balancing Clusters in Wireless Ad Hoc Networks. In Proceedings of the 3rd IEEE Symposium on Application-Specific Systems and Software Engineering Technology (ASSET'00), pp 25.

[4] D. J. Baker, and A. Ephremides, November 1981. The Architectural Organization of a Mobile Radio Network via a Distributed Algorithm. IEEE Transactions on Communications, COM-29(11): pp 1694– 1701.

[5] J. Broch, D. Maltz, D. Johnson, Y. Hu, and J. Jetcheva, October 1998. A performance comparison of multi-hop wireless ad hoc routing protocols. In Proceedings of the Fourth Annual ACM/IEEE International Conference on Mobile Computing and Networking.

[6] M. Denko, 2003The Use of Mobile Agents for Clustering in Mobile Ad Hoc Networks, In Proceedings of ACM International Conference, Vol. 47, pp 241 – 247.

[7] Z. J. Haas, and M. R. Perlman, The performance of query control schemes for the zone routing protocol. In Proceedings of ACM Sigcomm’98, Vol. 28, No. 4, pp 167 – 177, October 1998. J. Computer Sci., 2 (7): 583-588, 2006. 588.

[8] Z. J. Haas, and M. R. Pearlman, 2000. “The zone routing protocol: A hybrid framework for routing in ad hoc networks,” in Ad Hoc Networks, C. E. Perkins, Ed., Chapter 7, Addison-Wesley.

[9] I. Er, and W. Seah, July 2005. Clustering Overhead and Convergence Time Analysis of the Mobility-Based Multi-Hop Clustering Algorithm for Mobile Ad Hoc Networks, ICPADS, the 11th International Conference on Parallel and Distributed Systems.

[10] I. Er, and W. Seah, March 2004. Mobility-based d-Hop Clustering Algorithm for Mobile Ad Hoc Networks. In Proceedings of WCNC’04,

[11] M. Gerla, and J. Tsai, 1995. Multicluster, Mobile, Multimedia Radio Network. ACM Journal on Wireless Networks, Vol. 1, No. 3, pp 255- 265,

[12] A. McDonald, T. F. Znati, Aug. 1999. A mobility based framework for adaptive clustering in wireless ad hoc networks. IEEE JSAC, 17(8):1466- 1486,

[13] F. G. Nocetti, J. S. Gonzalez, and I. Stojmenovic, 2003. Connectivity Based k-Hop Clustering in Wireless Networks, Telecommunication Systems Journal, Vol. 22, No 1-4, pp 205-220.

[14] S. Sivavakeesar, and G. Pavlou, Stable Clustering Through Mobility Prediction for Large-Scale Multihop Intelligent Ad Hoc Networks, In Proceedings of the IEEE Wireless Communications and Networking Conference (WCNC'04), Georgia, USA, Mar. 2004, vol. 3, 1488-1493.

[15] Y. Wang, W. Wang, X. Li, 2005. Distributed Low-Cost Backbone Formation for Wireless Ad Hoc Networks, In the Proceedings of the 6th ACM international symposium on Mobile ad hoc networking and computing, pp 2 –13,

[16] Rainer Baumann, Simon Heimlicher, Martin May, ‘‘Towards Realistic Mobility Models for Vehicular Ad-hoc Networks’’ in 26th Annual IEEE Conference on Computer Communications, pp. 23-27, May 23-25, 2007.

[17] Uyen Trang Nguyen and Xing Xiong, ‘‘Rate-adaptive Multicast in Mobile Ad-hoc Networks’’, in IEEE International Conference on Wireless And Mobile Computing, Networking And Communications, pp. 352-360, Aug. 22-24,2005.

[18] Ammar Zahary and Aladdin Ayesh, ‘‘Analytical Study to Detect Threshold Number of Efficient Routes in Multipath AODV Extensions’’, in International Conference on Computer Engineering & System, pp. 95-100, Nov 27-29, 2007.

320325324