Joint Time-Frequency Anti-Jamming Communications: A Reinforcement Learning Approach

This paper investigates the channel selection and transmission duration scheduling problem in jamming environment. Although stable communication frequency and long transmission time can reduce switching overhead and achieve higher throughput, it is more likely to be disturbed by jammer. Our goal is to find the optimal transmission channel and duration strategy in jamming environment to maximize the long-term cumulative throughput of the system. We formulate the decision-making problem as a Markov decision process (MDP). Then, we propose a reinforcement learning (Q-learning) based channel selection and transmission duration scheduling algorithm. The simulation results show that compared with the reinforcement learning based fixed transmission duration algorithm, the system utility is significantly improved.