Low-Complexity Transceiver Design for Multi-User Millimeter Wave Communication Systems under Imperfect CSI

In this paper, we present a robust analog-digital hybrid transceiver design for millimeter wave (mmWave) communication systems operating in a K-user MIMO interference channel. The precoder design is based on minimizing sum-mean square error (sum-MSE) under the assumption that the transceivers possess only imperfect information about the channel state. The channel state information (CSI) error is assumed to follow Gaussian distribution. In order to reduce the hardware complexity of transceivers resulting from use of large number of antennas in mmWave systems, we propose reduced-complexity design using sparse approximation technique. We show that the proposed algorithm converges to a limit even though the global convergence is hard to prove due to the non-convex nature of the overall design problem. Furthermore, we evaluate the performance of proposed algorithm with various dictionaries. We numerically compare the performance of the proposed robust design with existing non-robust designs and the results demonstrate the resilience of the proposed algorithm to errors in the CSI.