Available online at www.sciencedirect.com ScienceDirect Procedia Computer Science 46 (2015 ) 1209 1215 International Conference on Information and Communication Technologies (ICICT 2014) Improving the Routing Performance of Mobile Ad hoc Networks Using Domination Set Preetha K G a, *, A Unnikrishnan a a Rajagiri School of Engineering & Technology, Kochi, 682039, India Abstract Mobile Ad hoc network is a self organized, self maintained network. Node movement is the important factor of increasing the control overhead in the network. The main objective of the paper is to reduce the control overhead by using the domination set based routing. The nodes which use to connect all the other nodes in the network are called dominating nodes, and the set of dominating nodes forms domination set. This paper proposes a new approach for finding the route and reducing the reroute establishment delay and increasing the packet delivery ratio. The efficiency of the method is demonstrated through simulation study. 2015 2014 The The Authors. Published by by Elsevier B.V. B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of organizing committee of the International Conference on Information and Communication Peer-review Technologies under (ICICT responsibility 2014). of organizing committee of the International Conference on Information and Communication Technologies (ICICT 2014) Keywords:MANET;AODV;DSR;Dominating node;dominating set;dbr 1. Introduction The evolution of mobile ad hoc networks (MANET) is growing in a rapid pace. Anytime anywhere networking came into picture because of the infrastructure less property of MANET. The nodes in the MANET are selfsufficient and each node is act as either router or source. There is no central controller in MANET 1 and the control is distributed among the nodes. Topology of the network is dynamic and the topology change is more frequent. Conventional routing algorithms are not suitable for MANET. There are many routing algorithms proposed for * Corresponding author. Tel.: 9387516291. E-mail address : preetha_kg@rajagiritech.ac.in 1877-0509 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of organizing committee of the International Conference on Information and Communication Technologies (ICICT 2014) doi:10.1016/j.procs.2015.01.035
1210 K.G. Preetha and A. Unnikrishnan / Procedia Computer Science 46 ( 2015 ) 1209 1215 MANET timely. Mainly the routing algorithms are categorized into two ie, proactive and reactive 6. Proactive algorithms are the extension of wired routing protocols. In this, all the nodes keep the routing information in the routing table, when the network is initialized. This information is periodically exchanged with its neighbors. Because of the periodic exchange, the routing overhead is very high in this case. In the reactive routing protocol on the other hand, the route is determined only when a node wants to send some data. In this case, there is some delay occurs for route establishment since the routing information is not readily available; many control packets are used for finding the route. The control packets also induce routing overhead in the network. As per study, it has been reported that the reactive algorithms are more efficient than proactive ones 8. The main constraints in MANET include high mobility, low bandwidth and low energy. Due to the high mobility, frequent disconnections are more in MANET. In all reactive algorithms, when the route is broken then the route re-establishment process diminish the performance of the network by inducing more overhead. This paper proposes a method to reduce the reroute establishment delay and routing overhead by using the dominating nodes in the network. Domination based routing came into existence 3,4 for a decade but no qualitative and quantitative analysis have been reported properly. In this work a scenario of Domination Based Routing is created and compared the performance against the existing algorithms. This clearly justifies that the proposed work is more efficient in terms of packet delivery ratio, control overhead and packet drop. In this method, initially all the dominating nodes in the networks are located. From this a domination set is created. A set is a dominating set, if all the nodes in the network are either in the set or the neighbors of the nodes in the set 2. The route is established through the members in the domination set only. All nodes in the network can be reached through the dominating nodes. When the route fails, it is easy to find the new route by using the domination nodes. This ensures the re-route establishment without any delay and overhead, thereby enhancing the routing performance, even when the route breaks occurs. The rest of the paper is organized as follows. Section 2 gives the brief idea about how the domination set is computed. Section 3 describes the proposed domination based routing. Simulation study and results are described in section 4 and the conclusion is given in section 5. 2. Computation of Domination Set In graph theory, the domination set is the subset of the graph such that each node is either in the set or has a neighbor in the set. The determination of the domination set is distributed among the nodes. Every node in the network can be reached through the domination nodes. Fig.1. An example mobile ad hoc network
K.G. Preetha and A. Unnikrishnan / Procedia Computer Science 46 ( 2015 ) 1209 1215 1211 Fig. 1. gives a typical example of mobile ad hoc network and in this the domination nodes are N3, N6 and N10. The domination set is not unique but the number of nodes in the set must be the same in all cases 2. 2.1Algorithm for finding domination set The algorithm below establishes the domination set by computing the adjacency matrix and collecting the nodes corresponding to the maximum connected node, from therow sum. Algorithm Node (X); If ( Y is a neighbour) { add to neighbour list send this list to its neighbours } If (X is a neighbour of Y) { adj[x][y]=1 else adj[x][y]=0 } For each row in adjacency matrix { //Find adjacency matrix //Compute the domination matrix } Compute the row sum in adjacency matrix Find the maximum connected node Append this to dom set If any node is not connected to the nodes in the dom set then add this also to the dom set 3. Domination Set Based Routing First step of the algorithm is to find the domination set. Then the route is established to the destination through the domination nodes only. The nodes in the domination set are able to connect all the nodes in the network very quickly. So it is easy to get the destination within no time. When the route failure occurs then the corresponding domination node identifies the problem and rectifies it locally. It can reach the destination through other nodes if possible. Otherwise it will flood the route failure report to the other domination nodes. In the initial phase, the domination nodes are determined form the adjacency matrix. For that each node determines its neighbour node by sending the HELLO packet. After determining the neighbours, the neighbouring list is sent to the adjacent nodes and each node prepares the adjacency matrix. From this matrix, it is easy to find out the dominating nodes and finally domination set, as by using the above mentioned algorithm in section 2.1. In Fig. 2. the domination set is <N3,N6,N10>. N1 is the source and N10 is the destination nodes. N10 can be reached from N1 through N3 and N6. In the initial route discovery process each node tries to connect the domination node from the source node. Even if the shortest link exists the route to destination is always through the dominating nodes. In the initial phase, the cost of establishing the route is higher than the other approaches; but this is very
1212 K.G. Preetha and A. Unnikrishnan / Procedia Computer Science 46 ( 2015 ) 1209 1215 useful in the dynamic networks, because mobile ad hoc network is highly dynamic in nature. So this is ready to lend a hand for finding the alternate route easily. In the given example if link N6-N8 is broken, then the node N6 can easily set up the connection N6-N7 to the destination. 4. Results and Discussions Fig. 2. Route discovery through dominating nodes The proposed algorithm DBR is implemented using NS2 simulator. The performance of DBR is compared with the existing algorithms AODV and DSR. It has been observed that DBR perform better than AODV and DSR in terms of packet delivery ratio, number of packets dropped and the routing overhead. The simulation parameters are listed in the Table. 1. Table.1. Simulation Parameters Parameter Value Routing Protocols AODV, DSR,DBR MAC Protocols IEEE 802.15.4 Number of Nodes 10,15,25,50 Simulation Area 500 X 500 Packet Size 512 bytes Simulation Time 500 seconds Packet delivery ration (PDR)= Network Overhead = When the route failure is detected in AODV or DSR, this will be reported by the intermediate node and the source node reinitiates the route discovery process. This will surely degrade the performance and throughput because some of the packet missed in the midst of the transmission. This issue can be rectified in DBR by introducing the local repair of route failure. The domination nodes are responsible for finding the alternate route to destination. Through the dominant nodes all the nodes in the network can be reached. Number of packet missed can be significantly reduced as shown in the Fig. 3. and the time to find out the route is also reduced. Fig.4. and Fig. 5.show the comparison of three algorithms in terms of packet delivery ration and routing overhead. In the first case,
K.G. Preetha and A. Unnikrishnan / Procedia Computer Science 46 ( 2015 ) 1209 1215 1213 the packet drop is rapidly increases in AODV and DSR, when the number of node increases. But in DBR the packet drop is not significant. Packet Delivery Ratio is almost similar in all algorithms. In DBR the routing overhead is reduced when the number of node increased; because if any link failure occurs, it is very easy to rectify in DBR. The efficiency of this approach is limited by the complexity of computation of domination set. Fig. 3. Number of packets dropped during transmission vs Number of Nodes; The DBR approach, proposed in the paper fairs better than the other two. Fig. 4. Packet Delivery Ratio vs Number of Nodes; PDR is highest for DBR, with larger number of nodes
1214 K.G. Preetha and A. Unnikrishnan / Procedia Computer Science 46 ( 2015 ) 1209 1215 4. Conclusion Fig. 5. Routing Overhead vs Number of Nodes; Overhead is consistently low for DBR. This paper explains the route discovery process in MANET and analyze the problem occurs during the route reestablishment process. When the route failure occurs the route establishment process starts from the beginning in the normal case. This causes many packet losses and the retransmission of the lost packet is high. But in domination set based routing (DBR) the route reestablishment, proposed here, is carried out by dominant nodes and the delay gets reduced. Accordingly new route discovery process is not required, while the active communication can be continued. This ensures the performance enhancement in terms of packet delivery ratio (PDR), number of packets dropped and the network overhead. The proposed algorithm works well in large networks also. However the determination of domination set, though required only occasionally could be a matter of concern in respect of overall computational complexity. References 1. Hong Jiang, Garcia-Luna-Aceves, J.J. Performance comparison of three routing protocols for ad hoc networks. Proceedings of the tenth International Conference on Computer communications and Networks, 2001.p. 547-554. 2. JIE WU,HAILAN LI. A Dominating-Set-Based Routing Schemein Ad Hoc Wireless Networks,Telecommunication Systems,2001;18: 13 36. 3. Jie Wu. Extended Dominating-Set-Based Routingin Ad Hoc Wireless Networks withunidirectional Links, IEEE transactions on parallel and distributed systems,2002;13: 9. 4. Jie Wu, Fei Dai, Ming Gao, Ivan Stojmenovic. On Calculating Power-Aware Connected Dominating Sets for Efficient Routing in Ad Hoc Wireless Networks, Journal of Communications And Networks, 2002; 4:1 5. Matulya Bansal,Gautam Barua. Performance Comparison of Two On Demand Routing Protocols for Mobile Ad hoc Networks,IEEE Personal Communication, 2002. 6. M. Abolhasan, T. Wysocki, E. Dutkiewicz. A review of routing protocols for mobile ad hoc networks, Ad Hoc Networks, 2004; 2:1 22. 7. Puneet Kumar Bhardwaj, Shipra Sharma,Vandana Dubey. Comparative analysis of reactive and proactive protocol of mobile ad hoc networks,international Journal on Computer Science and Engineering,2012; 4: 07. 8. SA Ade,PA.Tijare.Performance comparison of AODV, DSDV, OLSR and DSR routing protocols in mobile ad-hoc networks, International Journal of Information Technology and Knowledge management, 2010;2: 545-548. 9. CE. Perkins, EM. Royer,S. Das. Ad-Hoc On Demand Distance Vector Routing, RFC 3561, 2003. 10. Mehran Abolhasan, Tadeusz Wysocki, Eryk Dutkiewicz. A Review of routing protocols for mobile ad hoc networks, Journal of Ad hoc Networks,2004;2: 1-22.
K.G. Preetha and A. Unnikrishnan / Procedia Computer Science 46 ( 2015 ) 1209 1215 1215 11. G. Rajkumar, K. Duraisamy. A Review of AODV routing protocol for mobile ad hoc networks,journal of Theoretical and Applied Information Technology,2012; 36: 1. 12. Espes, D,Teyssie, C. Approach for Reducing Control Packets in AODV-Based MANETs, ECUMN, 2007.p. 93 104. 13. SA. Ade,PA. Tijare. Performance comparison of AODV, DSDV, OLSR and DSR routing protocols in mobile ad-hoc networks, International Journal of Information Technology and Knowledge management, 2010;2: 545-548. 14. Feng Xian,Jianzhi Wang. An Efficient Ad Hoc On-Demand Distance Vector Routing Protocol, IEEE Intl. Conference on Educational & Network Technology,2010. 15. Hui Xu, JJ. Garcia-Luna-Aceves. Efficient broadcast for wireless ad hoc networks with a realistic physical layer, Ad Hoc Networks 8: 165 180. 16. GS. Tomar. Modified Routing Algorithm for AODV in Constrained Conditions, Second Asia International Conference on Modelling & Simulation, 2008.