User-Centric Cell-Free Massive MIMO with Interference Cancellation and Local ZF Downlink Precoding

In the user-centric (UC) cell-free Massive MIMO architecture, a large number of distributed access points (APs) are deployed in a given area, each one serving the mobile stations (MSs) that are received with the largest power. This architecture has been shown to offer some advantages over a co-located massive MIMO architecture, especially in terms of fairness among users. This paper proposes to improve the performance of such a system by using local partial zero-forcing (PZF) precoding on the downlink and successive interference cancellation (SIC) on the uplink. The proposed schemes can be implemented locally, i.e. the channel estimates and the beamformers at the APs are computed and exploited locally, with no need to exchange information with a central processing unit (CPU); moreover the beamformers used at the MSs are channel independent, i.e. no channel estimate is needed at the MS. The numerical results show that the proposed transceiver algorithms provide remarkable performance improvements with respect to competing alternatives.