Adaptive Array Partitioning: A Novel Approach to Capturing Spatial Non-Stationarity in XL-MIMO

With the advent of extremely large-scale multipleinput multiple-output (XL-MIMO) systems, wireless channels exhibit challenging characteristics distinct from conventional systems, such as spherical wave effects and spatial non-stationarity (SnS). To mitigate the adverse impact of these characteristics on subsequent communication processes (e.g., channel estimation), this paper extends the near-field channel model for line-of-sight (LoS) XL-MIMO systems to accommodate SnS properties. First, we derive in detail the negative impacts of over-segmentation and under-segmentation under the influence of SnS, emphasizing the importance of precise subarray partitioning. Then, we introduce a power-adaptive subarray segmentation algorithm (PASS). Simulation results validate the accuracy of the proposed algorithm.