Direct Tensor-Based Estimation of Broadband mmWave Channels With RIS

We address the channel estimation problem in millimeter-wave (mmWave) systems aided by reconfigurable intelligent surface (RIS) and propose a peak detection-message passing (PDMP) algorithm. The angle and delay parameters of the channel are estimated from the multi-dimensional structure of the received tensor by exploiting the properties of the array steering vector (ASV). The channel gain is further estimated by employing the message passing method. The numerical results illustrate the efficiency of the proposed estimation method, especially in the low signal-to-noise ratio (SNR) regime.