Successive Interference Cancellation via Rank-Reduced Maximum A Posteriori Detection

This paper proposes a codeword-based iterative detection and decoding (IDD) algorithm for multiple-input multiple-output (MIMO) systems. In the proposed algorithm, multiple streams in a codeword are jointly detected at the rank-reduced (RR) maximum a posteriori (MAP) receiver and inter-stream interference is mitigated with successive interference cancelation (SIC). Thus, the algorithm is abbreviated to RR-MAP-SIC. Recent wireless standards such as Long-Term Evolution require the system to encode data bits per codeword, not per stream. As a result, conventional SIC algorithms could lose joint information among streams in a codeword because different streams of the same codeword are treated as interference. Instead, the proposed RR-MAP-SIC minimizes the loss of joint information by using the rank-reduced MAP detector. In addition, this paper compares the detector complexity of RR-MAP-SIC and investigates how the probability of symbol error is changed in terms of the covariance of the residual interference. As the number of iterations increases, the covariance is reduced so that the error events also decrease. Lastly, the extrinsic information transfer (EXIT) chart is used to analyze the performance of RR-MAP-SIC. Simulation results demonstrate the superiority of RR-MAP-SIC over the conventional algorithm and numerically verify the EXIT chart analysis.