Title:Hybrid Near-field and Far-field Localization with Holographic MIMO

Abstract:Due to its ability to precisely control wireless beams, holographic multiple-input multiple-output (HMIMO) is expected to be a promising solution to achieve high-accuracy localization. However, as the scale of HMIMO increases to improve beam control capability, the corresponding near-field (NF) region expands, indicating that users may exist in both NF and far-field (FF) regions with different electromagnetic transmission characteristics. As a result, existing methods for pure NF or FF localization are no longer applicable. We consider a hybrid NF and FF localization scenario in this paper, where a base station (BS) locates multiple users in both NF and FF regions with the aid of a reconfigurable intelligent surface (RIS), which is a low-cost implementation of HMIMO. In such a scenario, it is difficult to locate the users and optimize the RIS phase shifts because whether the location of the user is in the NF or FF region is unknown, and the channels of different users are coupled. To tackle this challenge, we propose a RIS-enabled localization method that searches the users in both NF and FF regions and tackles the coupling issue by jointly estimating all user locations. We derive the localization error bound by considering the channel coupling and propose an RIS phase shift optimization algorithm that minimizes the derived bound. Simulations show the effectiveness of the proposed method and demonstrate the performance gain compared to pure NF and FF techniques.