Transmitting over Ill-conditioned MIMO channels: from spatial to constellation multiplexing

This chapter addresses the problem of multiplexing multiple data streams in a multiple input multiple output (MIMO) system in the presence of channel matrix ill-conditioning brought by fading correlation and/or a Rice component. Conventional multiplexing schemes based on the separation of the stream's spatial signatures (commonly referred to as spatial multiplexing -SM, V-BLAST 1) rely on the linear independence between the channel responses corresponding to each transmit antenna. Consequently such schemes suffer considerably from effects bringing ill-conditioning in the MIMO channel matrix, such as fading correlation and Rice components. In an attempt to robustify SM schemes for deployment in in a wide range or propagation terrains, we investigate the use of socalled constellation multiplexing (CM) whereby distinct M-QAM streams are superposed to form a higher-order QAM constellation with rate equivalent to the sum of rates of all original streams. CM schemes do not rely on MIMO channel full rankness to function properly. We thus seek an approach that allows bridging SM and CM schemes. We show that this can be realized in the form of a linear diagonal precoder. This in turn yields an adaptive rate-preserving MIMO multiplexing algorithm that can operate smoothly at any degree of correlation or Ricean factor. Conventional SM and CM schemes are shown to be particular cases of the presented family of schemes.