System Level Performance of Downlink MU-MIMO Transmission for 3GPP LTE-Advanced

We investigate enhanced downlink multiuser multiple input multiple output (MU-MIMO) transmission schemes for 3GPP LTE-Advanced based on system level simulations. As compared to Rel. 8 LTE single user MIMO (SU-MIMO), MU-MIMO has the potential to better exploit the spatial dimension of MIMO channels in some scenarios. Although Rel. 8 LTE defines a simple MU-MIMO mode, more advanced MU-MIMO techniques are currently discussed towards Rel. 10 LTE-Advanced to further improve the spectral efficiency. In this paper, we focus on zero-forcing (ZF) MU-MIMO precoding with an explicit channel quantization mechanism called channel vector quantization (CVQ). Moreover, we consider two different criteria of CVQ at the receiver: either minimizing Euclidean distance or maximizing the signal to interference plus noise power ratio (SINR). A number of system level simulations reveal significant potential gains of ZF MU-MIMO as compared to those of Rel. 8 LTE SU/MU-MIMO in terms of both cell throughput and cell-edge user throughput when the spatial correlations at the base station (eNB) are relatively high. The potential gain of ZF MU-MIMO is shown to be further improved when the CVQ is based on the SINR maximization criterion instead of Euclidean distance minimization.