Sparse Bayesian Learning Based Symbol Detection for Generalised Spatial Modulation in Large-Scale MIMO Systems

Generalised spatial modulation (GSM) is extended from the concept of spatial modulation (SM). Due to its superior energy efficiency to classical multiple-input multiple-output (MIMO) techniques, GSM has attracted plenty of interest under the study of large-scale MIMO systems. In this paper, we propose a low-complexity symbol detector for GSM based on the framework of sparse Bayesian learning (SBL), which effectively recovers the GSM symbols by exploiting the inherent sparsity property of GSM. Compared to other sparsity-based symbol detectors, e.g., detectors based on the basis pursuit (BP) method and the orthogonal matching pursuit (OMP), the SBL-based detector is capable of achieving a superior reconstruction accuracy using less receive antennas. Numerical simulations are performed to substantiate the performance of the proposed detector.