Title:RIS-Assisted Passive Localization (RAPL): An Efficient Zero-Overhead Framework Using Conditional Sample Mean

Abstract:Reconfigurable Intelligent Surface (RIS) has been recognized as a promising solution for enhancing localization accuracy. Traditional RIS-based localization methods typically rely on prior channel knowledge, beam scanning, and pilot-based assistance. These approaches often result in substantial energy and computational overhead, and require real-time coordination between the base station (BS) and the RIS. To address these challenges, in this work, we move beyond conventional methods and introduce a novel data-driven, multiple RISs-assisted passive localization approach (RAPL). The proposed method includes two stages, the angle-of-directions (AoDs) between the RISs and the user is estimated by using the conditional sample mean in the first stage, and then the user's position is determined based on the estimated multiple AoD pairs in the second stage. This approach only utilizes the existing communication signals between the user and the BS, relying solely on the measurement of received signal power at each BS antenna for a set of randomly generated phase shifts across all RISs. Moreover, by obviating the need for real-time RIS phase shift optimization or user-to-BS pilot transmissions, the method introduces no additional communication overhead, making it highly suitable for deployment in real-world networks. The proposed scheme is then extended to multi-RIS scenarios considering both parallel and cascaded RIS topologies. Numerical results show that the proposed RAPL improves localization accuracy while significantly reducing energy and signaling overhead compared to conventional methods.