On Massive Spatial Modulation MIMO: Spectral Efficiency Analysis and Optimal System Design

Abstract

The idea of combining spatial modulation (SM) with massive multiple-input multiple-output (MIMO) system, i.e. massive SM-MIMO, has recently attracted plenty of academic interest, as it could simultaneously harness the advantages of SM and massive MIMO. However, the spectral efficiency (SE) analysis for massive SM-MIMO is still lack of exploring. To this end, a novel SE expression is derived in this paper to provide a tight lower bound for massive SM-MIMO systems, in which the impacts of imperfect channel estimation and random user locations are all accounted for. The new SE expression enables an efficient system-level analysis, via which we can investigate the optimal number of user antennas for SE maximization. Interestingly, we found that massive SM-MIMO maintains a higher SE than the conventional massive MIMO with single-antenna users when a limited number of users are scheduled, while conventional massive MIMO is more suitable when a large number of users are served.