Compressed Sensing for Channel State Information (CSI) Feedback in MIMO Broadcast Channels

摘要

In this paper, we proposed a new method to compress the CSI feedback. When the channel matrix is correlated, the DCT matrix works as a sparsifying basis to transform the channel matrix into a sparse form; the sparse signal is a feedback to the transmitter and reconstructed via the subspace pursuit (SP) recovery algorithm. Both theoretical analyses and simulation results show that the new method can introduce a huge computation cost reduction compared with the OMP algorithm and the codebook-based feedback scheme.