Pruning Stack Algorithm and Complexity Analysis

Stuck algorithm (SA) has been proved to be the most efficient one among the existing sphere decoding algorithms, although it needs a large amount of time-consuming comparisons. In this paper, a pruning stack algorithm (PSA) is proposed to decrease the size of the border node list in SA by statistical pruning. With a proper configuration, a good tradeoff is achieved between the performance and complexity. For high order modulations, PSA is able to reduce the number of comparisons by at least 30% as compared to SA without much penalty of the performance.