“One Plus One is Greater Than Two”: Defeating Intelligent Dynamic Jamming with Collaborative Multi-agent Reinforcement Learning

In this paper, we investigate the problem of anti-jamming communication in multi-user scenarios. The Markov game framework is introduced to model and analyze the anti-jamming problem, and a joint multi-agent anti-jamming algorithm (JMAA) is proposed to obtain the optimal anti-jamming strategy. In intelligent dynamic jamming environment, the JMAA adopts multi-agent reinforcement learning (MARL) to make on-line channel selection, which can effectively tackle the external malicious jamming and avoid the internal mutual interference among users. The simulation results show that the proposed JMAA is superior to the frequency-hopping based method, the sensing-based method and the independent Q-learning method.