Uplink-to-Downlink Covariance Matrix Estimation in ULA-equipped FDD Massive MIMO Systems

In this paper, we propose a novel downlink channel covariance matrix estimation scheme for massive multiple-input multiple-output systems with a base station equipped with a uniform linear antenna array (ULA). This scheme can be expressed in the form of easy-to-implement affine transformation which depends only on the uplink and downlink carrier frequencies, and the ULA configuration. In order to analyze the performance of the proposed scheme, we derive an upper bound on the estimation error. This bound shows that the accuracy of the proposed scheme increases with the number ULA antennas, the compactness and differentiability class of the periodic extension of a non-linearly transformed version of the angular power spread. Performance superiority of the proposed scheme over their existing counterparts is confirmed through numerical simulations.