A Low ML-decoding Complexity, Full-diversity, Full-rate MIMO Precoder

摘要

Precoding for multiple-input, multiple-output (MIMO) antenna systems isconsidered with perfect channel knowledge available at both the transmitter andthe receiver. For 2 transmit antennas and QAM constellations, an approximatelyoptimal (with respect to the minimum Euclidean distance between points in thereceived signal space) real-valued precoder based on the singular valuedecomposition (SVD) of the channel is proposed, and it is shown to offer amaximum-likelihood (ML)-decoding complexity of $\mathcal{O}(\sqrt{M})$ forsquare $M$-QAM. The proposed precoder is obtainable easily for arbitrary QAMconstellations, unlike the known complex-valued optimal precoder by Collin etal. for 2 transmit antennas, which is in existence for 4-QAM alone with anML-decoding complexity of $\mathcal{O}(M\sqrt{M})$ (M=4) and is extremely hardto obtain for larger QAM constellations. The proposed precoder's loss in errorperformance for 4-QAM in comparison with the complex-valued optimal precoder isonly marginal. Our precoding scheme is extended to higher number of transmitantennas on the lines of the E-$d_{min}$ precoder for 4-QAM by Vrigneau et al.which is an extension of the complex-valued optimal precoder for 4-QAM.Compared with the recently proposed $X-$ and $Y-$precoders, the errorperformance of our precoder is significantly better. It is shown that ourprecoder provides full-diversity for QAM constellations and this is supportedby simulation plots of the word error probability for $2\times2$, $4\times4$and $8\times8$ systems.