Semidefinite relaxation for detection of 16-QAM signaling

Abstract—We develop a computationally efficient approxima-tion of the maximum likelihood (ML) detector for 16 quadra-ture amplitude modulation (16-QAM) in multiple-input mul-tiple-output (MIMO) systems. The detector is based on a convex relaxation of the ML problem. The resulting optimization is a semidefinite program that can be solved in polynomial time with respect to the number of inputs in the system. Simulation results in a random MIMO system show that the proposed algorithm outperforms the conventional decorrelator detector by about 2.5 dB at high signal-to-noise ratios. Index Terms—Maximum likelihood detection, MIMO systems, semidefinite relaxation. I