Transmission Method of Electric Power Communication Network Based on Physical Layer Security

Based on the security problem in the electric power communication network, this paper studies the secure transmission method of the physical layer in an electric power communication network assisted by an intelligent reflecting surface (IRS). Considering the constraint of transmit power and the phase shift, the transmitted beam and the reflected phase shift of IRS are designed to maximize the total achievable safety rate. We combine model-free meta-reinforcement learning with the strategy gradient to solve the question of slow convergence and poor generalization. Simulation results reveal that the proposed method can achieve better performance than the convex optimization method and ordinary reinforcement learning.