A Constrained Clustering-Based Blind Detector for Spatial Modulation

In this letter, we propose a novel clustering-based detector for spatial modulation multiple-input multiple-output (MIMO) system. Specifically, we first convert unconstrained optimization problem of conventional K-means algorithm to the constrained optimization by controlling the number of received symbols in each cluster. Then, a low-complexity greedy algorithm is designed for solving the constrained optimization problem to determine the cluster centroids of the proposed detector and a novel detector based on the greedy algorithm is proposed accordingly. Simulation results show that the proposed detector efficiently avoids the occurrence of error floor effects of conventional K-means detector and can achieve near-maximum likelihood (ML) performance even if the number of clusters is large while effectively reducing the complexity compared to existing blind detectors.