A Low-Complexity MAP–SIC Detector for Massive MIMO Systems

In this paper, we propose a low-complexity maximum a posteriori detector with successive interference cancellation (MAP–SIC) for massive multiple-input multiple-output (MIMO) systems. The basic idea of the proposed MAP–SIC algorithm is to detect and cancel the signal of each user iteratively in order of approximate a posteriori log-likelihood-ratios (LLRs). To obtain the approximate a posteriori LLRs, the proposed method begins with the output of a matched-filter and estimates the mean and variance of interference-plus-noise term for each undetected symbol using the idea of Gaussian approximation. In addition, we also propose a simple strategy based on a posteriori log-likelihood-ratios (LLRs) to update the mean and variance in the iterative process of proposed algorithm. Since there is no need for a matrix inversion and exponential calculations, the complexity of the proposed detector is reduced significantly compared to that of minimum mean squared error (MMSE) and MMSE–SIC. Simulation results substantiate the performance of the proposed detector in the massive MIMO system.