Sensing-Assisted Compressive Superimposed CSI Feedback

In frequency division duplexing (FDD) massive multiple-input and multiple-output (mMIMO) systems, compressive sensing (CS)-based superimposed channel state information (CSI) feedback still faces significant challenges, such as the high computational complexity, the unavoidable channel estimation (CE) errors at the user equipment (UE), and superimposed interference. To address these challenges, we propose a sensing-assisted compressive superimposed CSI feedback method. By leveraging active sensing to extract the support set of downlink transmission paths, a sensing-assisted low-complexity reconstruction method is developed. In this method, the correctness verification and index supplementation schemes are employed to ensure the correctness of the sensed support set. Moreover, the CE error and superimposed interference are suppressed, thereby improving the recovery accuracy of both downlink CSI and uplink data sequence (UL-DS) while reducing computational complexity. Experiment results demonstrate that the proposed method substantially reduces the computational complexity. Furthermore, the proposed method presents performance improvement and robustness against parameter variations.