Reconfigurable Intelligence Surface Assisted Millimeter Wave MIMO Channel Estimation
In the reconfigurable smart surface(RIS)assisted millimeter-wave multiple-input multiple-output(MIMO)system,a channel estimation algorithm based on compressed sensing and deep learning is proposed for the sparsity of channel matrix introduced by millimeter-wave and the high-dimensional channel matrix introduced by RIS.Firstly,a small number of RIS components are linked to the RF link to construct a partial active RIS component-assisted millimeter-wave MIMO communication system.Sec-ondly,using the common sparsity of the angular domain between different subcarriers,the channel matrix is reconstructed by the sparsity adaptive matching(SAMP)algorithm and regarded as a two-dimensional noisy image.Finally,the symmetric two-way denoising network(STDN)is used to denoise the two-di-mensional noisy image.Among them,STDN is composed of two parallel symmetric paths,which can ex-tract features at different scales,remove image noise,and integrate the two feature maps to output the de-noising channel matrix.The simulation results show that the proposed SAMP-STDN algorithm improves the channel estimation accuracy.
万方数据
维普
