Hybrid preprocessing aided spatial modulation in multi-user massive MIMO systems

Massive multiple-input multiple-output (MIMO) as a promising technology to achieve high spectral efficiency encounters the problem of high cost and complexity of hardware implementation for digital signal processing. One proposed solution is hybrid analogue-digital processing to reduce the number of radio-frequency (RF) chains required at massive antenna arrays. In this study, hybrid analogue-digital preprocessing aided spatial modulation is introduced to the downlink of multi-user massive MIMO systems. Linear zero-forcing (ZF) and regularised ZF preprocessors which are near-optimal in massive MIMO systems and a low-complexity two-stage preprocessing scheme based on block-diagonalisation are proposed. These full-digital schemes are implemented in hybrid RF-baseband domain through a low-dimensional processor based on the baseband effective channel and phase controllers at the RF domain. The spatially correlated and sparsely scattered millimetre wave channels are adopted to consider the proposed preprocessing schemes. Under an imperfect channel state information at the transmitter, the robust design of full digital and hybrid analogue-digital preprocessors is presented using stochastic robust approximation. Furthermore, the hybrid RF-baseband preprocessors are realised using heavily quantised angles and the preprocessing schemes are investigated from the data rate point of view. The simulation results show desirable performance and close to that of full digital preprocessing is achieved by proper configuration of system.