An improved scheme of initial radius selection in SD algorithm under the low SNR

Abstract

Sphere decoding (SD) is an efficient algorithm for Multiple-input Multiple-output (MIMO) digital communications. It has been showed to achieve near Maximum Likelihood (ML) performance with low complexity. However, the complexity of conventional SD algorithm is high under the specific environment. The performance and the computation complexity of SD algorithm is highly dependent on the initial choice of sphere radius. Thus, it is important to design a proper initial radius scheme for SD which improve the performance and reduce the computation complexity. In this paper, we propose a novel scheme base on the knowledge of statistical probability and the constellation energy to design a new initial radius scheme. The performance and algorithmic efficiency of this scheme are increased by application of a factor constellation energy E. The simulation results show under the low signal-to-noise ratio (SNR) condition. The proposed scheme can achieve the aim that save more operation time than the conventional SD algorithm. At the same time, the performance gap compared to conventional SD algorithm can be restricted to a certain level.