Node stability-based routing in Wireless Mesh Networks

Network stability is a key performance metric in supporting real time communication over wireless networks. Because of high bandwidth demand and dynamic traffic variation, several paths in Wireless Mesh Networks (WMNs) are expected to be unstable. High levels of network instability can lead to interferences, packet losses and high delays. In this paper, we address the stability problem of WMNs. Instability in these networks is caused mainly by link quality fluctuations and frequent route flapping. First, we present the key factors that may cause network instability. Then, we propose a new technique, called Node Stability-based Routing (NSR), using the entropy function to define a node stability and a probability function to select an appropriate gateway. Simulation results show that NSR can significantly improve the overall network performance compared to techniques using interference and channel switching (MIC), Expected Transmission count (EIX) or load at gateways as a routing metric, Reinforcement learning-based best path to best gateway (RLBDR), and nearest gateway (i.e., shortest path to gateway).