Combined Rate and Precoder Design for Slow Fading Correlated MIMO Channels with Limited Feedback

The knowledge of channel state information at the transmitter (CSIT) is very important to enhance the spectral efficiency of MIMO links through precoder and power adaptation. In practice, only limited CSIT knowledge is usually available. In most existing works, the focus was on the precoder and power adaptation design for i.i.d. MIMO channels to optimize the ergodic capacity or SNR. The issues of rate adaptation and spatial correlation between antennas, which plays a key role in the overall system performance in slow fading channels, has been ignored in the existing works. In this paper, we shall propose an integrated framework for the rate and precoder adaptation under limited feedback for slow and spatially correlated MIMO fading channels. To capture the effects of potential packet errors, we consider average system goodput, which measures the b/s/Hz successfully delivered to the receiver, as our performance measure. The average system goodput is maximized subject to a total transmit power constraint and a target frame error rate (FER) constraint. Simulation results show our integrated design outperforms the Grassmannian precoder design in i.i.d. and spatially correlated MIMO channels. Furthermore, the limited feedback efficiency increases as the correlation coefficient increases.