On low complexity ML detection algorithm in MIMO system

An approximate maximum likelihood (ML) detection algorithm with low complexity for multiple input multiple output (MIMO) systems is proposed in this paper. Firstly, the traditional successive interference cancellation (SIC) method based on sorted QR decomposition is used to separate all the antennas into several groups and then only low complexity ML detection is performed in each group. Meanwhile, a great many complex multiplications would be replaced by complex additions and some low complexity real multiplications during each search step, and thereby more decoding time can be saved greatly. In order to overcome the performance bottleneck due to error propagation resulting from SIC operation, a new joint ML detection method among all groups is proposed to achieve the maximum diversity gain and the best error performance.