A multi‐agent reinforcement learning anti‐jamming method with partially overlapping channels

Abstract

This paper investigates the problem of multi-user anti-jamming channel access with partially overlapping channels (POC). Compared with traditional anti-jamming systems that use non-overlapping channels, POC improve the spectral efficiency. However, the partial overlap of channels also brings more serious interference. For this, physical distance and channel separation on the interference intensity under partially overlapping channels are first considered and the malicious jamming and interference among users are formulated as a hierarchical binary model. Secondly, to cope with multi-user decisions under dynamic jamming conditions, the Markov game framework is adopted to analyse the problem. Thirdly, a multi-user collaborative anti-jamming channel selection algorithm based on reinforcement learning is proposed as well as the optimal anti-jamming strategy can be obtained. Finally, the simulation results validate that the proposed algorithm helps users cope with jamming and eliminate mutual interference. Compared with the non-overlapping channel access scheme, the POC access scheme achieves higher network throughput.