RIS-Assisted Physical Layer Group Key Generation: A MARL Perspective

Physical-layer group key generation (PGKG) has been an attractive alternative for secure group communications in wireless networks. In this paper, we propose a reconfigurable intelligent surface (RIS)-assisted PGKG scheme for the first time, which introduces RIS technology to enhance the key generation performance via a hybrid configuration of random and optimized phase shifts. To address the coordination challenge between the RIS and nodes under decentralized deployment, we propose an algorithm based on a fully cooperative multiagent reinforcement learning (MARL) mechanism that jointly configures the RIS phase shifts and the nodes' broadcasting coefficients. Simulation results demonstrate that the proposed scheme outperforms existing works, achieving a higher sum secret key rate (SKR) along with stronger robustness against eavesdropping.