Maximizing Throughput using Adaptive Routing Based on Reinforcement Learning

In this paper, prioritized sweeping confidence based dual reinforcement learning based adaptive routing is studied. Routing is an emerging research area in wireless networks and needs more research due to emerging technologies such as wireless sensor network, ad hoc networks and network on chip. In addition, mobile ad hoc network suffers from various network issues such as dynamicity, mobility, data packets delay, high dropping ratio, large routing overhead, less throughput and so on. Conventional routing protocols based on distance vector or link state routing is not much suitable for mobile ad hoc network. All existing conventional routing protocols are based on shortest path routing, where the route having minimum number of hops is selected. Shortest path routing is non-adaptive routing algorithm that does not take care of traffic present on some popular routes of the network. In high traffic networks, route selection decision must be taken in real time and packets must be diverted on some alternate routes. In Prioritized sweeping method, optimization is carried out over confidence based dual reinforcement routing on mobile ad hoc network and path is selected based on the actual traffic present on the network at real time. Thus they guarantee the least delivery time to reach the packets to the destination. Analysis is done on 50 nodes MANET with random mobility and 50 nodes fixed grid network. Throughput is used to judge the performance of network. Analysis is done by varying the interval between the successive packets.