摘要

A new reduced algorithm based on the K-best sphere decoding and the successive interference cancellation algorithm is proposed for detection of multiple-input multiple-output systems. This algorithm divides the detection into two parts: the previous part detects by K-best SD algorithm, the later part detects by the SIC algorithm. Then a threshold based on the noise variance is proposed to further reduce the complexity of the algorithm. Any nodes those have metric larger than it will be cut during detecting. Simulation results show that the propose algorithm reduce the calculation complexity efficiently while at the same time the performance drop is almost neglectable.

全文