Reinforcement learning-based best path to best gateway scheme for wireless mesh networks

This paper addresses the problem of optimal routing in backbone wireless mesh networks (WMNs) where each mesh router (MR) is equipped with multiple radio interfaces and a subset of nodes serve as gateways to the Internet. Most routing schemes have been designed to reduce routing costs by optimizing one metric, e.g., hop count, load at routers and interference. However, when considering these metrics together, the complexity of the routing problem increases drastically. Thus, an efficient and adaptive routing scheme that takes into account several metrics simultaneously is needed. In this paper, we propose an efficient new routing scheme, called RLBPR (Reinforcement Learning-based Best Path Routing), that adaptively learns an optimal routing policy, depending on multiple optimization metrics such as loss ratio, interference ratio and load at the gateways. Simulation results show that RLBPR can significantly improve the overall network performance compared to schemes using either Metric of interference and channel switching (MIC), Best Path to Best Gateway (BP2BG), Expected Transmission count (ETX), nearest gateway (i.e., shortest path to gateway) or load at gateways as a metric for path selection.