Printer Friendly

A comparative study of mobile Ad hoc network protocols for throughput, average end-to-end delay and jitter.

Introduction

Scalable routing is one of the key challenges in designing and operating large scale Mobile Ad hoc Networks (MANET). In order to ensure effective operation as the total number of nodes in the MANET becomes very large, the overhead of the employed routing algorithms should be low and independent of the total number of nodes in MANET [1]. An important consideration in the development of scalable routing algorithms in large scale MANET is that the overhead properties of the scalable routing formally studied and analysed. In order for the ad hoc networks to operate as efficiently as possible, appropriate on-demand routing protocols have to be incorporated, to find efficient routes from a source to a destination, taking node mobility into consideration. The Mobility influences ongoing transmissions, since a mobile node that receives and forwards packets may move out of range. As a result, links fail over time. In such cases a new route must be established. Thus, a quick route recovery procedure should be one of the main characteristics of a routing protocol. It is also important to study the various performance metrics for better understanding and utilization of the routing protocols.

In this paper, we presented the results for various proactive and reactive routing protocols like Ad Hoc On-Demand Vector routing (AODV), Dynamic MANET On-demand (DYMO), Source Tree Adaptive Routing (STAR) protocol, Routing Information Protocol (RIP), Bellman Ford, LANd Mark Ad hoc Routing protocol (LANMAR) and Location Aided Routing protocol (LAR). The performance analysis of our study is restricted to throughput and delay metrics with and without concern of mobility.

Problem Formulations

AODV

The Ad Hoc On-Demand Distance Vector routing protocol (AODV) is an improvement of the Destination-Sequenced Distance Vector routing protocol (DSDV). It is based on distance vector and also uses the destination sequence numbers to determine the freshness of the routes. It operates on the On-demand fashion. AODV requires hosts to maintain only active routes. The advantage of AODV is that it tries to minimize the number of required broadcasts. It creates the routes on an on-demand basis, as opposed to maintain a complete list of routes for each destination. Therefore, the literature on AODV [2], classifies it as a pure on-demand route acquisition system. The usage of the AODV protocol for mobile ad hoc networking applications provided consistent results for large scale scenarios [3].

Dymo

The DYnamic MANET On-demand (DYMO) protocol is a reactive routing protocol being developed within IETF's MANET working group. Typically, all reactive routing protocols rely on the quick propagation of route request packets throughout the MANET to find routes between source and destination. While this process typically relies on broadcasting, route reply messages that are returned to the source rely on unicasting. DYMO is basically an improvement over the AODV protocol as for AODV every node records the next hop to send a packet to a specific destination [5].

RIP

Routing Information Protocol (RIP) is an Interior Gateway Protocol used to exchange routing information within a domain or autonomous system. RIP lets routers exchange information about destinations for the purpose of computing routes throughout the network. Destinations may be individual hosts, networks, or special destinations used to convey a default route. RIP is based on the Bellman-Ford or the distance-vector algorithm. This means RIP makes routing decisions based on the hop count between a router and a destination. RIP does not alter IP packets; it routes them based on destination address only.

Bellman Ford

Bellman-Ford Routing Algorithm, also known as Ford-Fulkerson Algorithm, is used as an algorithm by distance vector routing protocols such as RIP, BGP, ISO IDRP, and NOVELL IPX. Routers that use this algorithm will maintain the distance tables, which tell the distances and shortest path to sending packets to each node in the network. The information in the distance table is always updated by exchanging information with the neighbouring nodes. The number of data in the table equals to that of all nodes in networks (excluded itself). The columns of table represent the directly attached neighbours whereas the rows represent all destinations in the network. Each data contains the path for sending packets to each destination in the network and distance/or time to transmit on that path. The measurements in this algorithm are the number of hops, latency, the number of outgoing packets, etc.

LANMAR

LANMAR is an efficient routing protocol in a "flat" ad hoc wireless network [7, 8]. LANMAR assumes that the large scale ad hoc network is grouped into logical subnets in which the members have a commonality of interests and are likely to move as a "group" LANMAR uses the notion of landmarks to keep track of such logical subnets. It uses an approach similar to the landmark hierarchical routing proposed in [9] for wired networks. Each logical group has one node serving as landmark. The route to a landmark is propagated throughout the network using a Distance Vector mechanism [7]. The routing update exchange of LANMAR routing can be explained as follows.

Each node periodically exchanges topology information with its immediate neighbours. In each update, the node sends entries within its Fisheye scope [3]. Updates from each source are sequentially numbered. To the update, the source also piggybacks a distance vector of all landmarks. Through this exchange process, the table entries with larger sequence numbers replace the ones with smaller sequence numbers. As a result, each node has detailed topology information about nodes within its Fisheye scope and has a distance and routing vector to all landmarks. LANMAR outperform AODV protocol.

LAR

The Location--Aided Routing Protocol uses location information to reduce routing overhead of the ad-hoc network. Normally the LAR protocol uses the GPS (Global Positioning System) to get this location information. With the availability of GPS, the mobile hosts knows there physical location. To reduce the complexity of the protocol, we assume that every host knows his position exactly; the difference between the exact position and the calculated position of GPS will not be considered.

Design of the Experiment & Simulation Setup

The method for analyzing the routing protocols traffic is to begin with a carefully designed base configuration and network scenario for the experiment, and to vary the node density and mobility at a time to stress the network in different directions. Careful selection of these control parameters enables us to assess and isolate the effect of network size, with fixed application traffic CBR. In addition, design of the base condition, network topology, and routing are to be taken into account the real networks for which the results should be applicable.

In this experiment, we noted down the throughput, delay values and jitter for 30 nodes with random waypoint mobility of nodes for assessing the scalability issue for the routing protocols under consideration. In the beginning of the experiment, the initial settings of the node and simulation times were thoroughly checked out. Care also is taken in selection of the terrain dimension i.e. 1500m x 1500m. The experiment is continued for 30 nodes. In all these cases, we noted down the throughput, delay, jitter and real times of the simulator. We selected the terrain dimensions as 1500m x1500m cm, and nodes in the terrain are mobile. We fixed the simulation time for all the node densities and also varied according to the increments in node densities. The experiment used a static utilization of IPv4 networking protocol.

QualNet 4.5 is a scalable network simulation library that was designed with the primary goal of simulating large, high-fidelity models of wired, wireless, and mixed networks in an efficient manner. It was designed to achieve modular design for easy comparison of protocols under uniform conditions, detailed and accurate models, efficient execution, and transparent parallel execution for further scalability and runtime efficiency [10, 11]. These significant features encouraged us to use QualNet 4.5 (Academic version) for our study.

[FIGURE 1 OMITTED]

Mobility model

Nodes in the simulation set up move according to a model that is well known as the "random waypoint" model. The movement scenario files we used for each simulation are characterized by a pause time. Each node begins the simulation by remaining stationary for pause time seconds. It then selects a random destination in the 1500m x 1500m space and moves to that destination at a speed distributed uniformly between 0mps and a maximum speed of 10mps. Upon reaching the destination, the node pauses again for pause time seconds, selects another destination, and proceeds there as previously described, repeating this behaviour for the duration of the simulation. The simulation ran for 30 seconds of simulated time. We ran our simulations with movement patterns generated for a fixed pause time of 30 Seconds.

Application Traffic

As the goal of our simulation was to compare the performance of each routing protocol, we chose our application traffic sources to be constant bit rate (CBR) sources. When defining the parameters of the communication model, we experimented with sending rates of 1.2 packets per second and packet sizes of 512 bytes to observe the consistency.

Results and Discussion

Throughput

It is one of the dimensional parameters of the network which gives the fraction of the channel capacity used for useful transmission selects a destination at the beginning of the simulation i.e., information whether or not data packets correctly delivered to the destinations.

[FIGURE 2 OMITTED]

Average end to end delay

The average end-to-end delay of data packets is the interval between the data packet generation time and the time when the last bit arrives at the destination.

[FIGURE 3 OMITTED]

Jitter

The jitter is the variation of data communication packets in the network.

[FIGURE 4 OMITTED]

Throughput analysis

In the above experiment, we found that at node densities of 30. DYMO, LAR and RIP routing protocols showed higher throughput values. The results are compared with the extended results of already existing work [3] and found very much suitable for selecting routing protocols.

End to end delay analysis

As shown in Fig. 3, this simulation experiment showed us that AODV, DYMO and LAR protocols are having higher end to end delays than others, indicating that the speed of simulation in large scale networks will be affected by this. This analysis exclusively deals with the network speed and communication effectiveness. Higher the delay, lower is the speed and possibility of packet drop and so needs the fault tolerance approach of selecting these protocols.

Jitter analysis

As per the end-to-end delay, AODV, DYMO and LAR have higher jitter value and rest are having lower. Jitter determines the variation of networks.

Conclusion

In this paper, we compared the routing protocols based on significant performance metrics like throughput, delay and jitter. In this experiment we gone through some problems like communication stoppage for short durations; difference in simulation times for same scenario conditions (of course was solved by running the simulator for more than 10 times). We also faced the problem of switching off of the scenario for higher node densities. It might be due to the processor capability (RAM usage). We obtained the consistent results as compared with the literature [12, 13]. We believe that our work could be more intuitive for researchers for protocol selection and their suitability of application in real time scenario analysis in ad hoc networks.

References

[1] Yi Wang et. al, "Cluster based Location-Aware routing Protocol for Large Scale Heterogeneous MANET", in Proceeding of the Second International Multi symposium on Computer and Computational Sciences, IEEE Computer Society, 2007, pp.366-373.

[2] C. E. Perkins and E. M. Royer, "Ad-hoc on demand distance vector routing," in Proceedings of the 2nd IEEE Workshop on Mobile Computing Systems and Applications (WMCSA'99), vol. 3, New Orleans, LA, USA, February 1999, pp. 90-100.

[3] V.R. Sarma Dhulipala, RM. Chandrasekaran, R. Prabakaran, "Timing Analysis and Repeatability Issues of Mobile Adhoc Networking Application Traffics in Large Scale Scenarios", International Journal on Recent Trends in Engineering (IJRTE), Academy Publishers, Vol. 1, No. 1, May, 2009.

[4] L. Kleinrock and K. Stevens, "Fisheye: A Lenslike Computer Display Transformation," Technical report, UCLA, Computer Science Department, 1971.

[5] Marga Nacher et. al., "Multipath Extensions to the DYMO routing protocol", In Proceedings of the 9th International Conference on Mobile and Wireless Communications Networks, Cork, Ireland, September 19-21, IEEE, 2007.

[6] J. J. Garcia-Luna-Aceves and M. Spohn, "Source-Tree Routing in Wireless Networks," Proceedings of 7th International Conference on Network Protocols, 1999.

[7] M. Gerla, X. Hong, G. Pei, "Landmark Routing for Large Ad Hoc Wireless Networks," In Proceeding of IEEE GLOBECOM 2000, San Francisco, CA, Nov. 2000.

[8] G. Pei, M. Gerla, and X. Hong, "LANMAR: Landmark Routing for Large Scale Wireless Ad Hoc Networks with Group Mobility," In Proceeding of IEEE/ACM MobiHOC 2000, Boston, MA, Aug. 2000.

[9] P. F. Tsuchiya, "The Landmark Hierarchy: a new hierarchy for routing in very large networks," In Computer Communication Review, vol. 18, No. 4, Aug. 1988, pp. 35-42.

[10] http://www.scalable-networks.com

[11] http://www.qualnet.com

[12] J. Broch, D.A. Maltz, D.B. Johnson, Y.C. Hu, and J. Jetcheva, A Performance Comparison of Multi-Hop Wireless Ad Hoc Network Routing Protocols, Proc. of the ACM/IEEE International Conference on Mobile Computing and Networking (MobiCom), October 1998.

[13] M. Lakshmi and P.E. Sankaranarayanan, "Performance Analysis of Three Routing Protocols in Wireless Mobile Ad hoc Networks", Information Technology Journal, 5(1), pp.114-120, 2006.

Virendra Singh Kushwah (1), Kamal Kumar Chauhan (2) and Amit Kumar Singh Sanger (3)

ABV, Indian Institute of Information Technology & Management, Gwalior, India E-mail: (1) kushwah.virendra248@gmail.com, (2) kamalchauhans@gmail.com, (3) sanger.amit@gmail.com
Table 1: Simulation Setup.

Parameters Value

Simulation Area 1500m x 1500m
Number of nodes 30
Simulation duration 30 s
Routing protocol Bellman Ford, AODV, LANMAR, RIP,
 STAR, DYMO, LAR
Mobility pattern of nodes Random waypoint
Minimum motion speed 0 mps
Maximum motion speed 10 mps

Table 2: Throughput values of various routing protocols.

Routing Protocol Throughput (bits/second)

Bellman Ford 1296
LANMAR 1195
RIP 3117
AODV 2325
DYMO 4120
LAR 4100

Table 3: Average end to end delay values of various routing protocols.

Routing Protocol Average end to end delay (second)

Bellman Ford 0.03346
LANMAR 0.02336
RIP 0.02023
AODV 0.09933
DYMO 0.07791
LAR 0.08079

Table 4: Jitter values of various routing protocols.

Routing Protocol Jitter (second)

Bellman Ford 0.004203
LANMAR 0.005388
RIP 0.001582
AODV 0.03278
DYMO 0.04073
LAR 0.04291
COPYRIGHT 2010 Research India Publications
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2010 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Kushwah, Virendra Singh; Chauhan, Kamal Kumar; Sanger, Amit Kumar Singh
Publication:International Journal of Computational Intelligence Research
Article Type:Report
Date:Jul 1, 2010
Words:2400
Previous Article:Face recognition using geometric attributes.
Next Article:Java remote method invocation with GSS in distributed systems.
Topics:

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters