How to Leverage Double-Structured Sparsity of RIS Channels to Boost Physical Layer Authentication

Reconfigurable Intelligent Surfaces (RIS)-assisted systems are promising technology in next-generation wireless networks, but are susceptible to spoofing attacks due to their broadcast nature. This study reveals the unique characteristics of RIS-aided multiple-input multiple-output (MIMO) systems, that improve channel entropy compared to conventional MIMO. By capitalizing on the additional paths introduced by the cascaded channel and the distinctive double-structured sparsity inherent in its virtual representation, we develop a novel channel-based physical layer authentication (PLA) approach. In particular, we construct a robust signature for authentication purposes by extracting the intrinsic RIS features of the virtual angle of arrivals and departures indices. Furthermore, the distribution of the digital signature is analyzed to derive analytical expressions for the false alarm and detection probabilities of the proposed scheme. Simulation results show that the proposed approach surpasses the limitations of previous works, with 14.89% and 72% authentication performance improvements in detection and false alarm rates, respectively.