Channel Covariance Identification in FDD Massive MIMO Systems

Abstract

[Abstract]: Many channel estimation methods in Massive MIMO FDD systems usually rely on the knowledge of the channel covariance matrix to operate. However, in real scenarios, this covariance is not known beforehand, and hence it should also be estimated. In this work, we investigate different existing techniques for covariance identification to achieve full knowledge of this matrix with very short training sequences. Moreover, we propose a modification of the spatial smoothing approach with the goal of improving the quality of the channel covariance identification.