On the spectral efficiency of MMSE vector precoding

In this paper, we investigate the spectral efficiency of vector precoding with minimum mean square error (MMSE) linear preprocessing. We restrict the discussion to the spectral efficiency of MMSE vector precoding with quadrature phase-shift keying (QPSK) signaling. Spectral efficiency is investigated by numerical simulations, and plotted as a function of the energy per bit divided by the noise spectral density E_b/N₀. The optimum system load α, given as the ratio of the number of transmit and receive antennas, that maximizes spectral efficiency is obtained. Previously obtained spectral efficiency results for: Dirty paper coding (DPC), linear zero forcing (ZF), ZF vector precoding, and linear MMSE precoding are provided for comparison. We quantify the performance enhancement that MMSE vector precoding obtains in comparison to vector precoding with ZF linear preprocessing, in the low to medium E_b/N₀ region. We also find that MMSE vector precoding does not significantly outperform its linear counterpart.