On the ergodic capacity of jointly-correlated rician Fading MIMO channels

In this paper, we study the capacity-achieving input covariance matrices for the jointly-correlated (or the Weichsel-berger) Rician fading multiple-input multiple-output (MIMO) antenna channel when perfect channel state information (CSI) is known at the receiver while only statistical CSI at the transmitter is available. Unlike the Kronecker model, such jointly-correlated MIMO channel accounts for the correlation coupled between the two ends and has been shown to be most accurate for representing real channels. Our contribution includes the expression for the asymptotic mutual information for the jointly-correlated Rician fading MIMO channel in the large-system regime in which the numbers of antennas at the transmitter and receiver go to infinity with a fixed ratio. Based on this expression, an efficient algorithm is also proposed to obtain the capacity-achieving input covariance matrix. Simulation results demonstrate that even for a moderate number of antennas at each end, the proposed scheme provides undistinguishable results as those obtained by the highly-complex stochastic programming (or Monte-Carlo based) approach.