Joint Sparse Recovery Algorithm for Channel Estimation in XL-MIMO Systems

In Ultra-large-scale multiple-input multiple-output (XL-MIMO) systems, high-precision estimation of far-field channel state information (CSI) remains a core challenge for 6G high-speed transmission. Addressing the accuracy limitations of traditional orthogonal matching pursuit (OMP) algorithms caused by fixed grid partitioning, as well as the deficiencies of compressed sampling matching pursuit (CoSaMP) algorithms in capturing the common sparse structure of multi-antenna systems and their high computational complexity, this paper proposes a joint method combining improved CoSaMP with grid-free optimization. Specifically, we enhance sparse structure capture capability through a multi-signal residual correlation accumulation (MSRCA) mechanism, improve noise robustness via dynamic support set updates integrated with adaptive threshold strategies, and adopt continuous-domain grid-free optimization to directly estimate channel parameters, thereby avoiding grid mismatch. The simulation results show that the proposed scheme significantly outperforms existing methods in terms of channel estimation accuracy and pilot efficiency in far-field scenarios, while maintaining strong robustness under near-field interference, providing a highly adaptable solution for XL-MIMO systems.