Spectral efficiency of the in-band full-duplex massive multi-user multiple-input multiple-output system

In this study, the authors propose an in-band full-duplex massive multi-user multiple-input multiple-output system which exploits the separate-antenna arrays at base station and the shared-antenna at users. First, the channel state information of the two antenna arrays is estimated using the minimum mean-square-error method according to the channel reciprocity, and then the matched-filter and zero-forcing linear processing methods are adopted to analyse the system spectral efficiency (SE). The authors derive the lower and upper bounds of the uplink and downlink achievable rates in succinct forms to measure the system performance. The selection of full/half-duplex mode is also discussed with respect to the number of users. Numerical and simulation results show that the optimal downlink SE can be achieved with the increase of active users, and the full/half-duplex mode selection can help improve the system SE with the variation of traffic loads.