Semi-Blind Joint Channel Estimation and Symbol Detection for Multi-RIS Aided MIMO Systems

In multiple Reconfigurable Intelligent Surface (RIS)-assisted Multiple Input Multiple Output (MIMO) communication systems, the estimation of multi-hop cascaded channels poses significant challenges. This letter proposes a novel semi-blind receiver to address this problem originally. It shows that the received signals correspond to the generalized nested Parallel Factor (PARAFAC) model, which can be decomposed into three-dimensional (3D) PARAFAC models. We transform the joint channel estimation and symbol detection problem into several least square subproblems by deriving two cost functions for each 3D PARAFAC model, which are solved using the proposed SALS method. Simulation results demonstrate the effectiveness of the proposed algorithm.