Hybrid-Field Channel Estimation for RISs-Aided XL-MIMO Systems by Leveraging Signal Separation

This study focuses on providing accurate channel information to enable high-quality and high-capacity communication. Achieving precise channel state information (CSI) with low complexity is a key challenge, particularly in the terahertz band where shorter wavelengths exacerbate signal blockage and scattering caused by buildings. Reconfigurable intelligent surfaces (RIS) mitigate these issues by creating additional reflective paths, enhancing XL-MIMO system performance.In RISs-assisted XL-MIMO systems, hybrid-field communication arises as some users are in the far-field while others are in the near-field. To address this, the proposed Hybrid-field Stochastic Gradient Pursuit(HSGP) algorithm leverages the angular-domain sparsity of far-field channels and the polar-domain sparsity of near-field channels, using prior knowledge of channel path numbers. By incorporating row sparsity, HSGP efficiently extracts common paths and employs LMS to reduce computational complexity. Simulations validate the effectiveness of the method.