An Information Geometry Approach for Channel Estimation in Near-Field XL-MIMO Systems

Near-field channel estimation for extremely large-scale multiple-input multiple-output (XL-MIMO) systems can be formulated as obtaining the a posteriori information of the angle-distance domain channel. However, calculating the a posteriori mean and covariance is challenging owing to the sharp increase in antenna array and high resolution sampling in the angle-distance domain. To overcome this issue, we propose a near-field information geometry approach (NF-IGA) for channel estimation in XL-MIMO systems. Specifically, the angle and distance information of propagation environment is first obtained by the path parameters estimation layer. Then, a more lightweight polar domain dictionary is constructed for the information geometry layer, which reduces the complexity of calculating the a posteriori information. Simulation results verify the superiority of the proposed NF-IGA over other baseline methods.