Multipath Load Balancing : An Entropy based Clustering Solution for Mobile Ad hoc Networks

Mobile Ad-hoc networks are constrained by constantly changing topology, uncertain channel conditions and limited resources. Routing overhead increases with increased number of nodes and increased traffic. Multipath routing is popular technique used to improve the throughput and provide load balancing. To further enhance scalability and to achieve better quality of service ,mobile nodes can be grouped forming clusters to take advantage of the hierarchy in the network. . We propose a load balancing solution that distributes traffic across multiple paths for a cluster based routing protocol. Entropy based metric is applied to select cluster heads . Further entropy values of the cluster heads are extended to evaluate the stability of the link formed by the cluster heads. Simulation study show improved performance in terms of better cluster head lifetime and enhanced network performance.