Parameterized Separate Channel Reconstruction in RIS-Assisted Multi-User MIMO OFDM Systems

The channel estimation overhead problem is a key challenge for the reconfigurable intelligent surface (RIS)-assisted wireless communication systems. In this paper, we analyze the spatial consistency of multi-users for RIS-assisted multiple-input multiple-output orthogonal frequency division multiplexing (MIMO-OFDM) systems and classify users in each user group into two kinds, namely one central user (CUE) and several adjacent users (AUE). We propose an innovative two-phase multi-user joint parameter estimation (MJPE) scheme. In the first phase, we analyze the “orthogonality” of the CUEs in the UE-RIS channel and the “coherence” in the RIS-BS channel and propose a joint concurrent transmission (JCT) algorithm and a joint Newton orthogonal matching pursuit (JNOMP) algorithm, respectively. In the second phase, we propose an extrapolation algorithm for UE-RIS channel estimation based on the spatial proximity of adjacent users around the central users. In this scheme, separated channel parameters are estimated for each path for each user, including sparse paths for UE-RIS channels and non-line-of-sight (NLoS) paths for RIS-BS channel, based on known line-of-sight (LoS) path for RIS-BS channel. A new pilot protocol with an RIS phase control strategy is proposed, which can effectively reduce pilot overhead. In addition, we analyze the time computational complexity of the algorithm. The numerical results demonstrate that the proposed multi-user estimation algorithm can accurately accomplish channel reconstruction with low overhead.