IMPROVING THE QoS OF ONDEMAND MULTIPATH ROUTING PROTOCOL IN MOBILE AD HOC NETWORKS

Abstract

A Mobile Ad Hoc Network MANET is a continuously self configuring, infrastructure free network of mobile devices connected without wires newlineIn today s MANET, whether single or multipath routing protocol, delay in reaching a message, energy loss and weakening in the link quality occur newlineThere can be delay in reaching a message to destination from source and so the challenge is to minimize this delay newlineSince all nodes work on battery energy and since energy loss through a weakening battery is natural, the challenge is to make provision for sustained energy so that route failure does not occur newlineSince nodes may not be stable in one place, weakening of the link occurs, so the challenge is to maintain the stability of the link newlineIn order to minimize the End to End delay, we are proposing an algorithm entitled Delay based Ad Hoc on demand Multipath Distance Vector routing Protocol newlineIt takes care of the three processes in E2E delay, route discovery, route selection and route maintenance newlineIn the first, the rely nodes update the delay in route request RREQ packets, sent by the source node newlineThe destination node then calculates the E2E delay of entire route and sends the reply to the source node through route reply RREP message newlineThe source node then chooses the route based on this updated E2E delays which in turn reduces the overall delay considerably newlineThe best path chosen will beupdated in the routing table newlineWith regard to solving the energy loss, we have designed a new algorithm entitled Energy aware and delay based Ad Hoc On demand Multipath Distance Vector EAOMDV Routing protocol newlineAs for link stability, we use a procedure called Link Stability Energy Aware Routing protocol LSEA newline It is the extended version of EAOMDV routing protocol newlineThis ensures link stability of the network based on the movement of node newlineA comparative study between our algorithm with the existing ones have been done and the result clearly exhibits that ours algorithms are more efficient with respect of E2E delay, maximum remaining energy and link stability