Load eqUilibrium Neighbor Aware Routing in Mobile Ad Hoc Network

In Mobile Ad hoc Networks (MANETs), frequent link breakages occur which reduces the route lifetime and increases the frequency of invoking the route discovery process. Broadcasting is used in most of the reactive routing protocols for route discovery. Broadcasting increases routing overhead and end-to-end delay as well as decreases packet delivery ratio. To solve the broadcast storm and routing overhead problem, protocols with area based methods, probability based methods, neighbor knowledge based methods have been proposed. In protocols such as Probabilistic Counter-Based Route discovery (PCBR), Neighbor Coverage Dynamic Probabilistic Route Discovery (NCDP), Gossip-Based Ad hoc Routing (GBAR), Neighbor Coverage-Based Probabilistic Rebroadcast (NCPR) intermediate nodes carry out computation such as connectivity factor, rebroadcasting delay, rebroadcasting probability before rebroadcasting the route request packet which increase the overall delay in the network. Some of the protocols such as Load-Aware On demand Routing (LAOR), Associativity-Based Routing (ABR), node centrality based Load Balanced Routing (LBR) focus on load balancing using measurement of delay or traffic or both. However, in these protocols the intermediate nodes simply rebroadcast the route request packets to all the neighbors which increases the routing overhead of the network. In this paper, we propose Load eqUilibrium Neighbor Aware Routing (LUNAR), which combines the advantages of neighbor coverage knowledge and load balancing techniques to implement decision making system at every intermediate node. LUNAR significantly decreases the retransmission of route request packets and thus reduces the routing overhead within the network. We test our protocol using ns2 simulations of different scenarios. The scenarios have varying node density, node mobility, number of source-destination connections and queue length. The metrics we use for comparison with other protocols are packet delivery ratio, norm- lized routing overhead and end-to-end delay. Our results show LUNAR improve overall performance of the network by around 7-8 percentage, as compared to other routing protocols.