Tensor-based Channel Tracking for RIS-Empowered Multi-User MIMO Wireless Systems

The accurate estimation of Channel State Information (CSI) is of crucial importance for the successful operation of Multiple-Input Multiple-Output (MIMO) communication systems, especially in a Multi-User (MU) time-varying environment and when employing the emerging technology of Reconfigurable Intelligent Surfaces (RISs). Their predominantly passive nature renders the estimation of the channels involved in the user-RIS-base station link a quite challenging problem. Moreover, the time-varying nature of most of the realistic wireless channels drives up the cost of real-time channel tracking significantly, especially when RISs of massive size are deployed. In this paper, we develop a channel tracking scheme for the uplink of RIS-enabled MU MIMO systems in the presence of channel fading. The starting point is a tensor representation of the received signal and we rely on its PARAllel FACtor (PARAFAC) analysis to both get the initial estimate and track the channel time variation. Simulation results for various system settings are reported, which validate the feasibility and effectiveness of the proposed channel tracking approach.