Energy Efficient Position Based Opportunistic Routing Protocol for MANETS

Reliable and on-time data delivery is a crucial parameter of network performance in highly dynamic mobile ad hoc networks. Leveraging on a modified 802.11 MAC tweak, guard nodes relay the packet with prioritized back off time when the intended node fails and becomes a contender nodes themselves. Previously used ad hoc routing protocols of MANETS are susceptible to node mobility, especially for large-scale networks because of this tweak. To address this issue a MAC independent approach called Position-based Opportunistic Routing (POR) protocol uses location information to guide the data flow and can always archive near optimal path. Utilizing in-the-air backup mechanism, communication is maintained without being interrupted. On the other hand, this scheme focuses on the route discovery from the perspective of network layer and no complex MAC modification is necessary. Forwarding candidates are coordinated using the candidate list comparison available at every node . Although this method addresses the problem of node mobility and communication voids, this type of comparison is a severe drain on power resources of each node involved in the communication. So we propose to use POR in combination with Quorum-based asynchronous power-saving (QAPS) protocol. The basic idea of the proposed approach is to extend POR protocol for MANETs for saving more energy and accommodating more hosts. According to the proposed protocol, all the hosts of a locality are divided into clusters. Every cluster has one cluster leader with others being cluster members. The members are one-hop neighbours of the head and are synchronous with it. Prior approaches operated within an individual cluster. And the QAPS protocol is operated among cluster heads. All the cluster heads are organized as a virtual backbone to help route data. We perform simulations and compare the protocol with the related protocols. We find that the proposed protocol is more power-efficient and more scalable.