Reinforcement Learning based Space-Time-Frequency Domain Anti-jamming Transmission Scheme*

This paper applies reinforcement learning algorithms to optimize the anti-jamming strategies of transmitter, combining Multiple-input multiple-output (MIMO) communication systems with frequency hopping anti-jamming mechanisms. By perceiving the jamming channel information, the transmitting power and channel selection strategies are modeled as a Markov decision process. A Space-Time-Frequency Domain (STFD) anti-jamming transmission method based on the Win or Learn Fast-Policy Hill-Climbing (WoLF-PHC) algorithm is proposed to intelligently evade channel attacks. Simulation results validate that the proposed STFD anti-jamming method based on the algorithm WoLF-PHC outperforms the anti-jamming capability based on the Q-learning algorithm.