An Adaptive Threshold Channel Estimation Approach for RIS-Assisted Wireless Communications

The convergence of emerging technologies and consumer electronics is demanding for higher communication quality. Reconfigurable intelligent surface (RIS) can exhibit significant potential with its low cost and ease of deployment. However, the estimated channel accuracy is not satisfactory since channel state information is easily affected by the variation of environments in RIS-assisted wireless communication systems. To this end, an efficient channel estimation approach is proposed by using the adaptive threshold stage-wise orthogonal matching pursuit to extract the support atomic set and thus improve the estimation accuracy. Specifically, the modified particle swarm optimization algorithm is incorporated to select the optimal atomic threshold so that it can be dynamically adjusted to extract the support set efficiently according to the variation of the wireless channel, which is favorable for improving the channel estimation accuracy. In addition, the residual ratio threshold method is also introduced to optimize the iteration stopping condition, which is beneficial to the reduction of system complexity. Simulation results demonstrate that the proposed scheme can achieve a significant accuracy gain at least by 5% and reduce the pilot overhead at least by 33% as compared with some existing methods, which demonstrates the effectiveness of the proposed scheme in RIS channel estimation.