Low-complexity layered joint detection and decoding for LDPC coded large-MIMO systems

High performance and low complexity detectors are of great importance for MIMO systems, especially large MIMO systems, where there could be tens or hundreds of antennas. In this paper, we propose a layered factor graph based joint detection and decoding algorithm for LDPC coded large MIMO system. The joint factor graph which combines both detection and decoding is adopted, and Gaussian approximation is used to reduce the complexity. In the proposed algorithm, different from existing algorithms, in one iteration, the information of the variable nodes is updated one by one, instead of being updated as a whole. It is shown that the proposed algorithm can achieve better performance than the well-known MMSE based algorithm, but with much lower complexity. To reduce the complexity further, the half layered factor graph based algorithm and QRD pre-processing based algorithm are also proposed.