OMPL-SBL Algorithm for Intelligent Reflecting Surface-Aided mmWave Channel Estimation

Channel estimation (CE) is critical for intelligent reflecting surface (IRS) aided millimeter wave (mmWave) multiple input multiple output (MIMO) systems. In this paper, we propose the orthogonal matching pursuit list-sparse Bayesian learning (OMPL-SBL) algorithm which divides the cascaded channel estimation into two stages. The first stage calculates the prior for sparse Bayesian learning (SBL) using orthogonal matching pursuit list (OMPL) exploiting the grid sparsity of the cascaded channel, and the second stage employs the prior to obtain the accurate estimation using SBL. The proposed algorithm is able to achieve high estimation accuracy with low computational complexity compared to \boldmath ℓ1 -minimization and Bayesian algorithms. In simulation, we show the proposed algorithm not only cuts down time complexity by more than 95% of the SBL algorithm, but also achieves a higher estimation accuracy.