Resource prediction based routing using wavelet neural network in mobile ad-hoc networks

Many multimedia applications over Mobile Ad hoc NETworks (MANETs) require Quality of Service (QoS) to meet real-time services. Routing, an important component for QoS provisioning, affected by many factors such as limited resources, shared channel, unpredictable mobility, improper load balancing, and variation in signal strength. QoS based routing protocol must consider efficient usage of resources. In this paper, we propose a resource prediction based routing in MANET that routes the packets based on future availability of buffer-space, energy and bandwidth resources. Future availability of these resources is predicted using wavelet neural networks based traffic and mobility prediction model. The proposed routing scheme is simulated to evaluate the performance in terms of packet delivery ratio, computation overhead, memory overhead and packet delay.