Channel Estimation for Intelligent Reflecting Surface Assisted MIMO Systems: A Tensor Modeling Approach

Intelligent reflecting surface (IRS) is an emerging technology for future wireless communications including 5G and especially 6G. It consists of a large 2D array (semi-)passive scattering elements that control the electromagnetic properties radio-frequency waves so reflected signals add coherently at intended receiver or destructively to reduce co-channel interference. The promised gains IRS-assisted depend on accuracy channel state information. In this paper, we address design multiple-input multiple-output (MIMO) communication system via tensor modeling approach aiming estimation problem using supervised (pilot-assisted) methods. Considering structured time-domain pattern pilots IRS phase shifts, present two methods rely parallel factor (PARAFAC) received signals. first one has closed-form solution based Khatri-Rao factorization cascaded MIMO channel, by solving rank-1 matrix approximation problems, while second iterative alternating scheme. common feature both decoupling estimates involved matrices (base station-IRS IRS-user terminal), which provides performance enhancements in comparison competing are unstructured LS channel. Design recommendations guide choice parameters discussed. Numerical results show effectiveness proposed receivers, highlight trade-offs, corroborate their superior compared LS-based solutions.