RIS_CE

A multi-scale attention based channel estimation framework is proposed for reconfigurable intelligent surface (RIS) aided massive multiple-input multiple-output systems, in which both hardware imperfections and time-varying characteristics of cascaded channel are investigated. By exploiting the spatial correlations of different scales in the RIS reflection element domain, we construct a Laplacian pyramid attention network (LPAN) to realize the high-dimensional cascaded channel reconstruction with limited pilot overhead. In LPAN, we leverage the multi-scale supervision learning to progressively capture the spatial correlations of cascaded channel in a coarse-to-fine fashion, where the attention mechanism based dual-branch architecture is designed to fuse the high-frequency component and component of channel. To balance network performance and complexity of LPAN, we further propose a lightweight LPAN-L architecture. In LPAN-L, the partial standard convolutional layers are decomposed into the group convolution, dilated convolution and point-wise convolution, which forms a sparse convolutional filter set to extract the channel feature with less computation cost. In addition, we leverage parameter sharing and recursion strategy to reduce the required space complexity of LPAN. Furthermore, a selective fine-tuning strategy is developed to realize the domain adaption of the proposed LPAN-L model. Simulation results show that the proposed LPAN can achieve higher estimation accuracy than the existing estimation schemes, while the LPAN-L architecture with a close performance to LPAN reduces approximately half of the parameters and the computational cost