A DRL-Based Hierarchical Game for Physical Layer Security Aware Cooperative Communications
Denghui Liu, Ruoyang Chen, Tong Zhang, Changyan Yi
Abstract
This paper investigates a dynamic game framework for physical layer security (PLS)-aware wireless network with third-party collaborative users (TCUs). In the considered system, TCUs choose to assist legitimate users (LUs) in resisting eavesdropping attacks or opt to assist eavesdroppers (EVs) in eavesdropping on legitimate communication, in exchange for rewards from LUs and EVs. Due to the unpredictability of wireless systems, coalitions among the three parties may be dynamically changing. A hierarchical game integrating a matching sub-game with a coalition formation sub-game is formulated to model the interactions among LUs, EVs, and TCUs. In order to derive the optimal long-term strategies while solving the equilibrium of the hierarchical game, we develop a deep reinforcement learning (DRL)-based solution, which includes a matching process and a coalition selection process for two subgames. Simulation evaluates the performance of the proposed scheme and demonstrates its superiority over counterparts.