Increasing performance of ODMRP by using learning automata

Abstract

A Mobile ad hoc network is a collection of wireless communication nodes that dynamically form a network without any fixed infrastructure. One of the most important topics in these networks is multicast routing. Although multicast algorithms are desirable in many situations, their forwarding mechanism and network resource consumption makes them significantly less efficient than unicast routing algorithms. The most significant type of multicast protocols are mesh-based algorithms which, use periodic flooding mechanism to construct paths. Flooding rate in the multicast routing protocols depends on node mobility and input traffic. Thus, this rate should adaptively adjust to restrict overheads, while there are continuously a complete mesh between receivers and senders in network. In this paper, a novel prediction model is proposed to compute the best rate of flooding mechanism. Our results show substantial improvement in overhead while has no reduction in the performance of the multicast routing protocol.