一种低复杂度的MIMO预处理球形译码算法  

A Low Complexity Pre-processed Sphere Decoding Algorithm for MlMO Detection

作　　者：李庆坤[1] 马红光[1] 李正生[1] 左广霞[1] 

机构地区：[1]第二炮兵工程学院,陕西西安710025

出　　处：《信号处理》2009年第12期1867-1870,共4页Journal of Signal Processing

摘　　要：提出了一种新的低复杂度MIMO预处理球形译码算法。球形译码算法是MIMO译码方案达到最大似然性能的低复杂度译码算法,但当信噪比较低或调制阶数较高时,该算法的复杂度仍然很大,甚至接近穷尽搜索方法的复杂度。本文利用迫零译码的初始估值作为软信息,对信道矩阵各列进行重排列,从而改变解向量中各维信号的求解顺序,因此大大减少了低信噪比或较高调制阶数的球形译码计算复杂度。仿真结果表明,当Eb/No为0～3时,16QAM调制方式下的V-BLAST译码复杂度减少了近1/3。A new low complexity pre-processed sphere decoding algorithm for multiple-input multiple-output （MIMO） detection is proposed. The sphere decoding is a low complexity algorithm which may achieve the maximum-likelihood performance, but the complexity is still high when the Signal-to-Noise （SNR） is low or the modulation is in high order, and the complexity even approaches to the brutal search method. The columns of the channel matrix have been rearranged according to the soft information gotten from the zero- forcing decoding, so the order of the solution changed accordingly. This reduces the complexity of the sphere decoding greatly when the SNR is low or the modulation is high order. The result of simulation demonstrates that the complexity has been reduced nearly 1/3 when Eb/No in the range of 0 -3 and using the 16QAM modulation.

关 键 词：低复杂度 预处理 球形译码 

分 类 号：TP301.5[自动化与计算机技术—计算机系统结构]