Routing Optimization by Appending ACO With AODV in MANETS

Abstract

MANET is one of the most popular networks these days due to their feature of being temporarily installed and removed when the desired goal has been achieved. They provide a cost effective solution as they don’t require any predefined infrastructure i.e. all the nodes are autonomous in nature with no centralized control. Hence each node itself performs the function of host and router. Apart from this their self-configuring and self-healing nature allows them to implement in almost all fields (defence applications, home applications, education fields etc.). The main issue related to these networks is they require a routing algorithm which makes the network efficient by increasing the throughput rate and reducing average. One of the most efficient reactive routing algorithm named as AODV is being used to find the optimal path between source and destination node. The main advantage of using such algorithms is that nodes only need to track the network information when a request for transmission is being initiated. Unlike pro-active routing algorithms which includes updating nodes with each and every network information regardless of whether transmission is being taken place or not. A random network of nodes is deployed and the shortest path between the desired nodes is obtained using AODV routing protocol. Link failure occurs very frequently in networks such as MANETs due to mobility of nodes, dynamic network topology etc. Hence the routing protocol must be capable of handling such issues in an efficient manner. AODV has a drawback that local link repair is not possible; it shifts to an alternate route in such cases. This may lead performance degradation of the network. To avoid this it is necessary to find alternate route in case of link failure from the point of occurrence of failure. Hence an ACO (Ant Colony Optimization) appended approach has been proposed in to achieve this goal. ACO is a type of Swarm Intelligence technique inspired from the collective behavior of the biological ants in search for their food. ACO provides an efficient solution in finding routes with minimization of routing overheads in the network. In the proposed work the evaluation of ACO appended AODV algorithm is carried out on the basis of certain network parameters such as throughput, end-to-end delay, packet loss ration and route failures. The comparison of proposed approach with the traditional AODV approach revealed that the new algorithm outperformed the existing AODV approach in terms of network performance.