Joint Channel Estimation for RIS-Aided mmWave MIMO Wireless Communication Systems With Mixed-Resolution Quantization Schemes

Reconfigurable intelligent surfaces (RIS) are considered as a kind of potential technology of 6G communication, which can affect wireless communication channels by manipulating the reflection units. RIS is a kind of passive devices, which causes channel estimation to require large pilot overhead, and multiple-input multiple-output (MIMO) systems also bring huge power consumption and hardware cost. Therefore, we consider a RIS-aided mmWave MIMO communication system based on mixed-resolution quantization schemes to solve these problems. Considering the row and column sparse structure in the mmWave channel and the influence of mixed-resolution quantization, a channel estimation scheme based on the compressive sensing (CS) algorithm is proposed. However, if the Angles of Arrival (AoAs) or the Angles of Departure (AoDs) are not located on discrete grids, the sparse structure of angular channel matrices will leak. Taking full account of the particular leakage structure, a leakage structure orthogonal matching pursuit (LSOMP)-based channel estimation scheme is proposed. The simulation results show that the proposed preprocessing method can greatly improve the channel estimation accuracy, and the proposed algorithms with the mixed-resolution quantization schemes can not only reduce the circuit device power consumption of receivers, but also reduce pilot overhead and improve the performance stability especially in high SNR.