Secure Transmission Method of Power Communication Network based on Illegal Reconfigurable Intelligent Surface

Based on the security problem in electric power communication network, this paper studies the safe transmission method of electric power communication network based on illegal intelligent reflective surface. We studied IRIS-assisted eavesdropping scenarios, and jointly optimized transmitted beamforming and illegal beamforming to maximize the overall safety rate. We combine model-free meta-reinforcement learning with strategy gradient method to solve the problem of slow convergence and poor generalization. The simulation results show that the meta-reinforcement learning method proposed in this paper can effectively reduce the harm of IRIS, and compared with the traditional reinforcement learning algorithm, the algorithm can improve the total safety rate.