Secure Underwater Distributed Antenna Systems: A Multi-Agent Reinforcement Learning Approach

Dear Editor, Underwater distributed antenna systems (DAS) are stationary infrastructures consisting of multiple geographically elements (DAEs) which interconnected through high-rate backbone networks [1]. Compared to centralized systems, the DAS could provide a larger coverage area and higher throughput for underwater acoustic (UWA) transmissions. In this work, exploiting low sound speed in water, multi-agent reinforcement learning (MARL)-based approach is proposed secure against eavesdropping at physical layer. Specifically, theoretical secrecy rate firstly derived time-slotted UWA (UWANs) considering large propagation delays. Furthermore, we investigate long-term sum optimization problem under MARL framework, where each DAE learns its optimal transmission strategy online. Simulation results show that method achieves performance compared competing benchmark methods.