International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)

Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com Volume 2, Issue 1, January February 2013 ISSN 2278-6856

Performance Evaluation of AODV Routing Protocol in Wireless Sensor Networks with the Constraints of varying terrain areas by varying pause time
Siddharth Singh1, Dr.Naveen Hemrajani 2
1

Student M.Tech (IC), Department of Information Communication Suresh Gyan Vihar University Jaipur, Rajasthan, India
2

Professor, Department of CS/IT, Suresh Gyan Vihar University Jaipur, Rajasthan, India

Abstract:

A Network which have wireless sensor nodes animatedly form an communication-less network or provisional network without using existing network infrastructure is defined as wireless network .[1]The sensing of information in decisive conditions during the emergency state where the sensor network is deploy is its main importance. Benefit of developing sensor network is increasing for practical information in physical environment in different applications either manually or randomly. [2]Through this paper, we discuss and evaluate the performance of different network parameters on different topologies based on varying the pause time and keeping the speed constant (node speed) in different terrain areas which is small (1000 m. x 1000 m.), large (2000 m. x 1000 m.) and very large (2000 m. x 2000 m.) using AODV routing protocol and monitoring of critical conditions with the help of important parameters like Packet delivery Fraction, Average End- to- End Delay, Average Throughput, NRL and Packet loss. Keywords: Wireless Sensor Network, Packet Delivery

practically various scenarios are generated by fluctuating the pause time (node mobility) while keeping speed constant. [5] AODV is a routing protocol used for data transfer between mobile nodes. It transfers packets to nodes which they not directly communicate. It also uses the DSR routing protocol feature i.e. on-demand characteristic. In this paper we describe in Section 1 Introduction Section 2 Routing Protocol Section 3 Simulation Tool Section 4 Simulation parameters Section 5 Research Work Section 6 Simulation Setup Section 7 Results and Analysis and section 8 Conclusion.

2. EFFECTIVE ROUTING PROTOCOL

There is a lot of difference between DSR routing protocol and AODV routing protocol in maintaining routing information. AODV uses conventional routing tables for one entry per destination while DSR uses multiple route cache entries for each destination. At each target node AODV retains sequence number to prevent routing loop and to get newly routing information. [13] Neighbors: Two nodes are called neighbor nodes if they communicate directly with each other. By message each node can have a track of their neighbor node. In AODV routing, if a node want to communicate with a not neighbor node, it apply a route request message (RREQ), it consists several key information like the source, the destination, sequence number (Unique ID) etc. [14] Sequence number: The number which serves as a unique id is the sequence number. It allows nodes to match up updates in information by other nodes. When a node wants to send its own message, it gives a sequence number. Updates of information are hold by each node regarding all other nodes and higher sequence number defines which one has more updated route and more accurate information. Page 75

Fraction, End-to-End Delay, Network Simulator

1. INTRODUCTION
A group of small sensor nodes and wireless communication abilities are called wireless sensor networks (WSNs). Main advantages of WSNs are easy deployment, fast communication and low maintenance. WSNs are designed by using many routing protocols, power management protocols and data dissemination protocols so that energy can be saved. [2] WSNs are capable of sensing and making wireless communication of small nodes. [3] WSNs are mainly made up of sensors which are equipped with a radio and it forms a wireless network together. Its purpose are fast, easy deployment and low maintenance cost. [5] The purpose of this experimental study is to analyses the efficiency or capability and their behavior of changing network topology while continuing to deliver data packets successfully to their destination. To evaluate this work Volume 2, Issue 1 January - February 2013

International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)

Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com Volume 2, Issue 1, January February 2013 ISSN 2278-6856
Error Message: The significant characteristic of AODV is that it maintains time base category for individual routing tables of each node. When a route error message (RERR) is received by node, all the routes are checked by it that contains dreadful nodes and routing table of node. AODV maintain routes by the route error message and warns to each node when error message arise. [15] Features of AODV: (1) It finds routes when it required. (2)To check accuracy of update information of route it uses sequence number. (3) Only next hop for a route is defined by it instead of whole route. (4) It propels HELLO message to track neighbor. [16] to the total packets. Packet Loss [%] = (dropped Packets/ (total packets)) *100)

5. RESEARCH WORK
There are many research papers on routing protocols in wireless sensor network and all are used for evaluating performance of different parameters in different scenario. Researchers specify the difference between routing protocols and its performance for different parameters and which one is best for the case of Wireless Sensor Network. In comparison of AODV, DSDV and DSR the Average end-to-end delay and throughput in DSR are very high. While in comparison of DSDV and AODV routing protocols, AODV performed better than DSDV in terms of bandwidth as AODV do not contain routing tables so it has less overhead and consume less bandwidth while DSDV consumes more bandwidth. [12] In this paper we selected to investigated AODV protocol for different performance parameters for different Terrain areas like small (1000 m. x 1000 m.), large (2000 m. x 1000 m.) and very large (2000 m. x 2000 m.) based on varying pause time with keeping maximum node speed constant.. Analysis were done using ns-2 simulator on these three cases of terrain areas in order to derive an estimation of the performance parameters.

3. SIMULATION TOOL
NS-2 is an object based tool which encapsulates independent objects linked to each other within a system hierarchy. [7] Network simulators are very important and efficient analyzing tool used for routing in network and different protocols used for wired and wireless networks. [6] Network simulator use TCL to configure the topology, the nodes, the channels, schedules the events etc. C++ language is used to implements the protocols. [8] ns-2 is often used network simulator and is one of the most popular simulators for the researchers. ns-2 is extended to the wireless sensor network and its protocols. ns-2 uses object oriented design for implementation of different modules of a sensor network. [6]

6. SIMULATION SETUP
In this paper, we investigated AODV protocol with a scenario where a total of 100 nodes are used with the maximum connection number 10; CBR connection; transfer rate is taken as 4 packets per second i.e. the send rate of 0.25 and the pause time is varied starting from 0 s., 20 s., 40 s., 60 s., 80 s., and 100 s. (i.e. in the steps of 20 s.) implemented respectively in a 1000 m. x 1000 m., 2000 m. x 1000 m. and 2000 m. x 2000 m. terrain areas. The simulation time was taken as 100 seconds and keeping the speed constant. The details of general simulation parameter used are depicted in Table 1. Table 1: Simulation Parameter Values S. Parameters Values No. 1 Transmitter range 250m 2 Bandwidth 2 Mbps 3 Simulation time 100 sec 4 Number of nodes 100 5 Max Speed 10 6 Pause time 0, 20, 40, 60, 80, 100 sec 7 Terrain Area 1000 m. x 1000 m., 2000 m. x 1000 m. and 2000 m. x 2000 m. 8 Traffic type Constant Bit Rate 9 Packet size 512 bytes data 10 MAC type IEEE 802.11b 11 Antenna type Omni-Antenna Page 76

4. SIMULATION PARAMETERS
(I) Packet Delivery Fraction (PDF): Packet Delivery Fraction is the ratio of the number of data packets successfully delivered to the destination nodes and number of data packets produced by source nodes. [9] (II) End-to-End Delay: The term End-to-End delay refers to the time taken by a packet to be transmitted across a network from source node to destination node which includes retransmission delays at the MAC, transfer and propagation times and all possible delays at route discovery and route maintenance. [11] The queuing time can be caused by the network congestion or unavailability of valid routes. [17] (III) Throughput: The term throughput refers the number of packet arriving at the sink per ms. Throughput is also refers to the amount of data transfer from source mode to destination in a specified amount of time. The goal is to calculate maximum throughput of IEEE 802.11 technologies in the MAC layer for different parameters such as packet size. [10] (IV) Normalized Routing Load [%] (NRL): It is the number of routing packet required to be send per data packet delivered. NRL = (Number of Routing Packet) / (Number of Packet Received) (V) Packet Loss [%]: It is the number of dropped packet Volume 2, Issue 1 January - February 2013

International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)

Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com Volume 2, Issue 1, January February 2013 ISSN 2278-6856
12 Radio propagation method Two Ray Ground Speed = 10 m/s, Routing protocol = AODV, and Evaluating Average End-to-End Delay [in ms] Table 3: Evaluating Average End-to-End Delay [in ms] by varying pause time using AODV routing protocol
Pause Time Topology 1000 m. x 1000 m. 2000 m. x 1000 m. 2000 m. x 2000 m. 0 20 40 60 80 100

7. RESULT AND ANALYSIS

The investigations were performed on Parameters using AODV routing protocol such as Packet Delivery Fraction [%] (PDF), Average End-to-End Delay [in ms], Average Throughput [in kbps], Normalized Routing Load [%] (NRL) and Packet Loss [%]. The experimental data are shown in Tables 2 to 6 respectively and their respective performance being shown in Figure 1 to 5 respectively by Varying the Pause Time and keeping the Speed Constant = 10 m/s. When Nodes = 100, Pause Time = 0 - 100 s, Maximum Speed = 10 m/s, Routing protocol = AODV, and Evaluating PDF Table 2: Evaluating PDF by varying pause time using AODV routing protocol
Pause Time Topology 1000 m. x 1000 m. 2000 m. x 1000 m. 2000 m. x 2000 m. 0 98.8 8 84.9 9 79.1 5 20 94.7 3 90.9 5 44.6 2 40 98.3 9 95.7 8 59.3 7 60 98.3 5 93.6 5 16.3 8 80 99.3 7 96.4 8 68.2 8 100 99.5 5 97.0 2 32.6 6

49.72 1 313.1 3 448.9 1

115.9 9 123.2 1 480.8 9

49.96 0 49.75 7 431.5 2

40.27 9 168.0 9 1385. 5

43.15 8 82.43 0 77.38 3

31.61 0 97.97 9 38.98 9

Average End-End delay

1300 1200 1100 1000 900 800 700 600 500 400 300 200 100 0 0 20 40 60 80 100 Pause tim e (secs)

AODV-1000x1000 AODV-2000x1000 AODV-2000x2000

Packet Delivery Fraction (PDF)

100 90
Packet delivery fraction (%)

AODV-1000x1000 AODV-2000x1000 AODV-2000x2000

80 70 60 50 40 30 20 10 0 0 20 40 60 80 100 Pause time (se cs)

Fig. 2 Pause time versus Average End-to-End Delay [in ms]when terrain areas is 1000 m. x 1000 m., 2000 m. x 1000 m. and 2000 m. x 2000 m. by varying pause time using AODV routing protocol Like DSR, using AODV with 100 nodes, maximum Speed 10 m/s, varying pause time (0 - 100 s in intervals of 20 sec) for different terrain areas, we examine that Average End-to-End Delay [in ms] for small terrain areas is nearly constant (1000 m. x 1000 m.) and increases with larger terrain areas (2000 m. x 1000 m. and 2000 m. x 2000 m). In general, the Average End-to-End Delay increases for larger terrain areas. The Average End- toEnd Delay is however lesser than DSR for large terrain areas. Also, for WSNs, the Average End-to-End Delay [in ms] in general decrease with increased pause time. We can derive a formula according to simulation results as: Average End- to-End Delay Terrain Areas

Fig. 1 Pause time versus PDF when terrain areas are 1000 m. x 1000 m., 2000 m. x 1000 m. and 2000 m. x 2000 m. by varying pause time using AODV routing protocol Using AODV with 100 nodes, maximum Speed 10 m/s, varying pause time (0 - 100 s in intervals of 20 sec) for 1000 m. x 1000 m., 2000 m. x 1000 m. and 2000 m. x 2000 m. terrain areas, we examine that PDF for small terrain areas is nearly constant (1000 m. x 1000 m.) and decreases with larger terrain areas (2000 m. x 1000 m. and 2000 m. x 2000 m). In general, the PDF decreases for larger areas. We can derive a formula according to simulation results as: Packet Delivery Fraction 1/Terrain Areas

When Nodes = 100, Pause Time = 0 - 100 s, Maximum Speed = 10m/s, Routing protocol = AODV, and Evaluating Average Throughput [in kbps] Table 4: Evaluating Average Throughput by varying pause time using AODV routing protocol
Pause Time Topology 0 20 40 60 80 100

When Nodes = 100, Pause Time = 0 - 100 s, Maximum Volume 2, Issue 1 January - February 2013

Average delay (secs)

Page 77

International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)

Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com Volume 2, Issue 1, January February 2013 ISSN 2278-6856
1000 m. x 1000 m. 2000 m. x 1000 m. 2000 m. x 2000 m. 93.03 1 79.49 9 74.36 4 89.126 85.811 2 31.469 92.73 2 90.63 4 55.83 5 92.89 8 88.04 7 15.36 4 92.47 7 90.98 8 64.23 6 93.7 97 91.7 67 30.4 83

Results are same as DSR. In AODV the NRL is more for large terrain areas. NRL Terrain Areas When Nodes = 100, Pause Time = 0 - 100 s, Maximum Speed = 10m/s, Routing protocol = AODV, and Evaluating Packet Loss [%] Table 6: Evaluating Packet Loss by varying pause time using AODV routing protocol
Pause Time Topology 2000 m. x 1000 m. 2000 m. x 2000 m. 1000 m. x 1000 m. 0 20 40 60 80 100

Average Throughput [kbps] 100 90 80 70 60 50 40 30 20 10 0 0 20 40 60 80 100 Pause time (secs)

AODV-1000x1000 AODV-2000x1000 AODV-2000x2000

Average Throughput [kbps]

14.973 0 18.139 7 1.1170

8.965 53.39 9 5.272 5

2.08 9 38.5 0 1.56 1

5.54 8 81.8 8 1.55 9

3.392 8 28.19 5 0.632 3

2.804 9 62.41 5 0.446 2

Fig. 3 Pause time versus Average Throughput when terrain areas is 1000 m. x 1000 m., 2000 m. x 1000 m. and 2000 m. x 2000 m. by varying pause time using AODV routing protocol
Packet Loss [%]

Packet Loss [%]

AODV-1000x1000

100 90 80 70 60 50 40 30 20 10 0 0 20 40 60 80 100 Pause time (secs)

AODV-2000x1000 AODV-2000x2000

Using AODV with 100 nodes, maximum Speed 10 m/s, varying pause time (0 - 100 s in intervals of 20 sec) for 1000 m. x 1000 m., 2000 m. x 1000 m. and 2000 m. x 2000 m. terrain areas, we examine that Average Throughput are same as DSR. However, Average Throughput in AODV is less in contrast with DSR for large terrain areas. Average Throughput [kbps] 1/Terrain Areas When Nodes = 100, Pause Time = 0 - 100 s, Maximum Speed = 10m/s, Routing protocol = AODV, and Evaluating NRL Table 5: Evaluating NRL by varying pause time using AODV routing protocol
Pause Time Topology 1000 m. x 1000 m. 2000 m. x 1000 m. 2000 m. x 2000 m. 0 1.75 4.51 4.79 20 5.52 5.81 7 40 2.01 2.38 4.09 60 1.82 4.35 10.53 80 1.13 3.01 2.12 100 0.5 8 2.3 9 3.6 3

Fig. 5 Pause time versus Packet Loss when terrain areas are 1000 m. x 1000 m., 2000 m. x 1000 m. and 2000 m. x 2000 m. by varying pause time using AODV routing protocol Results are same as DSR. Here it may be noted that for medium terrain areas (2000 m. x 1000 m.) packet loss is higher in DSR, while nearly equal for large terrain areas (2000 m. x 2000 m.)

Packet Loss Terrain Areas

8. CONCLUSION
The results of our simulations are analyzed and discussed in this section. The results are analyzed and discussed in different terrain areas having networks of 100 sensor nodes on varying Pause time (00-100secs with interval of 20secs.) for evaluating performance of different parameters like Packet delivery Fraction, End- to- End Delay, Average Throughput, NRL and Packet loss in small, large and very large terrain areas. Our study provides an optimal result which is fully based on simulation and analysis. Every case explains evaluation of parameter with the help of table and generated graph. Each case represents a special issue for metric and Terrain area (which is small (1000 m. x 1000 m.), large (2000 m. x 1000 m.) and very large (2000 m. x Page 78

NRL 35.3 30.3 25.3

N RL

AODV-1000x1000 AODV-2000x1000 AODV-2000x2000

20.3 15.3 10.3 5.3 0.3 0 20 40 60 80 100 Pause time (secs)

Fig. 4 Pause time versus NRL when terrain areas is 1000 m. x 1000 m., 2000 m. x 1000 m. and 2000 m. x 2000 m. by varying pause time using AODV routing protocol Volume 2, Issue 1 January - February 2013

International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)

Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com Volume 2, Issue 1, January February 2013 ISSN 2278-6856
2000 m.)). According to the analysis value we drive a formula for each case that fully satisfies the values and relationship between parameters and terrain areas which is small (1000 m. x 1000 m.), large (2000 m. x 1000 m.) and very large (2000 m. x 2000 m.). The overall results says that when we implement sensor nodes in small terrain areas give better performance rather than Large and very large terrain areas. for Monitoring Applications, Masters Degree Thesis, Karlskrona October 2009. [12] Asar Ali Zeeshan Akbar, Evaluation of AODV and DSR Routing Protocols of Wireless Sensor Networks for Monitoring Applications, Masters Degree Thesis, Karlskrona October 2009 [13] IETF Manet Working Group AODV Draft, http://www.ietf.org/internet-drafts/draft-ietf-manetaodv-08.txt [14] Luke Klein-Berndt, A Quick Guide to AODV Routing, Wireless Communications Technologies Group, National Institute of Standards and Technology [15] Natarajan Meghanathan, Brajesh Kumar Kaushik, Dhinaharan Nagamalai Advances in Networks and Communications, First International Conference on Computer Science and Information Technology, CCSIT 2011, Bangalore, India, January 2-4, 2011. Proceedings, Part 2 [16] K.Thirunadana Sikamani, PK Kumaresan, M Kannan, R. Madhusudhanan, Simple Packet Forwarding & Loss Reduction for Improving Energy Efficient Routing Protocols in Mobile Ad-Hoc Networks, European Journal of Scientific Research, ISSN 1450-216X Vol.37 No.2 (2009), pp.254-260 [17] Dharm Singh, Chih-Heng Ke, Jagdish, Ankush Badala, Throughput Limits of IEEE 802.11b, IEEE 802.11b WG, September 1999. AUTHORS
Siddharth Singh pursuing M.Tech(Information Communication) from Suresh Gyan Vihar University, Jaipur (Raj), India and completed his Bachelor of Engineering in Computer Science from Rajasthan university, India 2008. His area of interests lies in the field of Operating System, Computer Architecture and WSN and research areas. He is working as an Assistant Professor in CSE/IT Department of JNIT College, Jaipur(Raj), India. He has three years Academic/Research experience. Prof. (Dr.) Naveen Hemrajani, Principal(Engg.), SGVU and Chairman CSI(Jaipur Chapter) received his B.E degree in Computer Science & Engineering from Shivaji University in the year 1992 and M.Tech(CSE) in 2004. His Research Topic for PhD was Admission Control for Video Transmission over IP Networks. He possesses 20 years of Teaching and research experience. He has published three books and many research papers in International and National Journals of repute. He has also presented several papers in International and National conferences. He is also Editorial Board member of many international Journals of repute. He is also working on DST(Department of Science & Tech.) sanctioned project.

REFERENCES
[1] Md. Anisur Rahman, Md. Shohidul Islam, Alex Talevski,Performance Measurement of Various Routing Protocols in Ad-hoc Network, Proceedings of the International MultiConference of Engineers and Computer Scientists 2009 Vol I, IMECS 2009, March 18 - 20, 2009, Hong Kong. [2] Ravindra Prakash Gupta, Dr. Virendra Kumar Sharma, Vijay Mohan Shrimal Investigation of Different Parameters of Dynamic Source Routing with varied Terrain Areas and Pause Time for Wireless Sensor Network, International Journal of Modern Engineering Research (IJMER) ISSN: 2249-6645, pp 626-631. [3] Jamal N. Al-Karaki Ahmed E. Kamal, Routing Techniques in Wireless Sensor Networks: A Survey, ICUBE initiative of Iowa State University, Ames, IA 50011. [4] J.M. van Dam, An Adaptive Energy-Efficient MAC Protocol for Wireless Sensor Networks, parallel en gedistribueerde systemen, June, 2003. [5] Vijay Mohan Shrimal, Ravindra Prakash Gupta, Virendra Kumar Sharma, Investigation of Adhoc Topology AODV for Wireless Sensor Networks for Varying Terrain Areas for Different Speed (node speed) International Journal of Advanced Research in Computer Science and Software Engineering ISSN: 2277 128X , Vol (2), No. (1) Jan 2012. [6] Kevin Fall, Kannan Varadhan, Editors, the VINT Project, UC Berkeley, LBL, USC/ISI, and Xerox PARC, the ns Manual. [7] The ns Manual, the VINT Project, Kevin Fall, Kannan Varadhan, May 9, 2010 [8] Imad Aad, Mohammad Hossein Manshaei, and JeanPierre Hubaux, ns2 for the impatient, EPFL Lausanne, Switzerland, March, 2009 [9] Siti Rahayu, Abdul Aziz, Nor Adora Endut, Shapina Abdullahand Mior Norazman Mior Daud,Performance Evaluation of AODV, DSR and Dymo Routing Protocol In Manet Conference on scientific & Social research, Cssr 0809, 14-15 March 2009. [10] Ian F. Akyildizy Mehmet C. Vurany O zgu r B. Akanz Weilian Su, Wireless Sensor Networks: A Survey Revisited, Computer Networks Journal (Elsevier Science), [11] Asar Ali, Zeeshan Akbar, Evaluation of AODV and DSR Routing Protocols of Wireless Sensor Networks Volume 2, Issue 1 January - February 2013

Page 79