Power Angular Spectrum and Spatial Covariance Estimation for Wideband mmWave RIS Channels

In this study, we propose a method for estimating cascaded channel covariance matrices (C-CCMs) for millimeterwave (mmWave) reconfigurable intelligent surface (RIS)-aided massive multiple-input multiple-output (MIMO) systems. Our approach utilizes a generic ray tracing channel model, wherein each multipath cluster (MPC) comprises several rays scattered randomly around a mean angle, following a specific probability density function (pdf). Initially, we estimate the mean angles of arrival (AoAs) at the base station (BS). Subsequently, we introduce a subspace-aware (SA) orthogonal matching pursuit (OMP) algorithm to estimate mean cascaded angles-delays-Doppler shifts. These estimated parameters are then used to construct subspace basis matrices, from which initial CCM estimates are derived. We further refine the initial CCM estimates by fitting pdfs to their eigenvalues. The effectiveness of our proposed methods is demonstrated through root mean square error (RMSE), normalized mean square error (NMSE), and power angular spectrum (PAS) metrics.