Sparsity-Aware Channel Estimation for Fully Passive RIS-Based Wireless Communications: Theory to Experiments

This article proposes a sparsity-aware channel estimation scheme for reconfigurable intelligent surface (RIS)-assisted wireless communications. We present an angular domain-channel sparsity model in a closed-form mathematical expression. A comprehensive formulation of the RIS channel estimation problem based on the sparsity analysis and RIS reflection model is also presented in this manuscript. This work aims to achieve a practical RIS beamforming algorithm without requiring excessive training overhead or any sensor deployment. We achieve the goal by proposing two sparsity-aware RIS channel estimation schemes based on the compressive sensing (CS) algorithms, such as the Dantzig selector (DS) and orthogonal matching pursuit (OMP). Differently from the existing works on the CS algorithms for RIS, we consider a fully passive RIS without any active sensor. We validate the theory and algorithm through both simulations and experiments. We have experimented orthogonal frequency-division multiplexing (OFDM) communications on our 5.8-GHz 1-bit RIS testbed with QPSK, 16QAM, 64QAM, and 256QAM modulation schemes. By experiments, it is shown that the proposed scheme is able to adaptively form a beam toward the receiver and improves the quality of the wireless communication to a notable level. Thanks to the properties of the proposed scheme, the wireless channel can be estimated without excessive training time and complexity.