Transmit Antenna Correlation in Spatial Multiplexing Systems

In this letter we present a transmit antenna correlation in spatial multiplexing systems with channel state information at the transmitter (CSIT) of a multiple-input multiple-output (MIMO) correlated channel. Indeed, high correlation degrades system capacity and performance. Most related work has thus far considered in the statistical channel model such as Kronecker model with an approximate Toeplitz form of the transmitter or receiver side correlation for measuring the Power Azimuth Spectral (PAS), but channel capacity has not been well studied. The Toeplitz assumption of the correlation matrix restricts the actual configuration of MIMO systems to a linear structure for a uniform linear array (ULA). Therefore, we only propose an optimal circulant assumption of a novel transmit correlation matrix in this letter to significantly reduce arithmetic complexity and spectral norms. We use this result to maximize ergodic capacity and minimize average bit error rate (BER) in spatial multiplexing (SM) systems with a zero-forcing (ZF) linear receiver. Simulation results show that the proposed method can achieve significant performance enhancement versus conventional equal power transmission.