Faster-than-Nyquist spatiotemporal symbol-level precoding in the downlink of multiuser MISO channels.

This paper investigates the problem of interference among the simultaneous multiuser transmissions in the downlink of multiple antennas systems. Symbol-level precoding (SLP) is a promising technique which has recently demonstrated large performance gains over the conventional block-level techniques. These gains can be translated in lower power requirements, improved energy efficiency, lower peak to average power ratio and resilience to non-linearities. However, previous works have not exploited the full potentials of SLP as it was only used to exploit multiuser interference spatially. In this paper, we extend this concept by using Faster-than-Nyquist (FTN) signaling and employing SLP to manage both multi-user and inter-symbol interference (ISI). We consider the aforementioned paradigm in the context of Massive multiple-input multiple-output (MIMO) systems, where the number of transmit antennas is usually an order of magnitude larger than the number of served users. In this rich degrees of freedom (DoF) environment, we show that FTN SLP can double the effective user rates while improving the energy efficiency.