Improved Sorting Architecture for K-Best MIMO Detection

Abstract

This brief presents an improved sorting architecture for K-best multiple-input multiple-output (MIMO) detection. In the proposed architecture, two properties of the K-best MIMO detection are exploited for simplifications: the order among siblings from the same parent node can be easily identified; the order among the K best nodes is dispensable. Besides, the generalized mergesort-based architecture for the K-best sorting is presented so that the proposed sorter can be constructed for any power-of-two K and modulation scheme. Compared to the conventional ones, the proposed architecture lowers not only the hardware complexity but also the latency significantly.