Reinforcement Learning-based Homogeneous Coexistence Interference Management in Wireless Body Area Network

Abstract

Current trends in remote health monitoring to monetize on Internet of Things applications has given a rise in efficient and interference free communications in wireless body area networks (WBANs). Coexistence interference in WBANs has aggravates the over-congested radio bands, thereby requiring efficient carrier sense multiple access with collision avoidance (CSMA/CA) strategies to improve interference management. Existing solutions utilize simplistic heuristics to approach interference problems. The scope of our research work is to investigate reinforcement learning for efficient interference management under coexisting scenarios with an emphasis on homogenous interferences. We propose an intelligent CSMA/CA mechanism based on reinforcement learning called QIM-MAC efficiently utilize sense the slots with minimum interference. The simulation results show that our approach maximizes average network throughput and packet delivery ratio (PDR), and minimizes energy consumption, average delay, and packet loss ratio.