Channel Statistics based Adaptive Feedback for Cooperative Massive MIMO Systems

Abstract

Feedback overhead for channel state information (CSI) acquisition and accordingly backhaul overhead for base station (BS) cooperation can be significantly reduced by channel statistics based codebook in cooperative massive multiple-input multiple-output (MIMO) systems. When utilizing the skewed codebook based on the channel covariance matrices, the effect of the quantization error on the achievable rate depends on the channel covariance matrices of the serving and interfering channels. Leveraging the analysis, we propose channel statistics based adaptive feedback by deriving optimal feedback bit allocation to the serving and interfering channels in a closed-form according to their channel covariance matrices.