Distributed routing in Ad Hoc networks with some QoS constraints

Ad Hoc wireless networks consist of mobile nodes (hosts) interconnected by multihop communication paths or radio links. Due to the dynamic nature of Ad Hoc networks, the state information (delay and bandwidth) for each one of the mobile nodes is inherently imprecise. This paper presents a routing algorithm to deal with imprecise state information related to the end-to-end delay and end-to-end available bandwidth for every node within the network. The algorithm uses fuzzy inference systems to provide itself with an intelligent decision making engine that will be able to conduct search for paths that satisfy the quality of service requirements with the least delay and the highest bandwidth. The results showed high success ratios without flooding the network with routing messages.