一种改进的排序QR分解V-BLAST检测算法  

An improved sorted QR factorization based detection algorithmfor V-BLAST

作　　者：谢伟[1] 何培宇[1] 

机构地区：[1]四川大学电子信息学院,成都610064

出　　处：《四川大学学报（自然科学版）》2012年第4期817-822,共6页Journal of Sichuan University(Natural Science Edition)

基　　金：通信抗干扰技术国家重点实验室资助项目(JG2007055)

摘　　要：垂直分层空时编码(V-BLAST)是贝尔实验室提出的一种MIMO系统.在基于QR分解算法的基础上,结合最大似然算法,提出了一种改进算法,称为排序ML-QR算法.该算法先对信道矩阵按列范数从小到大进行列置换,再进行QR分解,对最先检测的两层信号采取最大似然检测,提高最先检测级的性能,剩下层的信号采用QR分解算法检测.理论分析及仿真实验结果表明,提出的排序ML-QR算法优于传统的分层空时检测算法,在高信噪比的情况下性能改善更加明显.Vertical layered space-time coding （V-BLAST） is a MIMO System Proposed by Bell Labs. Combining QR factorization algorithm and maximum likelihood （ML） algorithm,an improved algorithm called Sorted ML-QR algorithm is proposed. At first the channel matrix is sorted from its smallest ele- ment to the largest by column norm~ then the channel matrix is factorized with QR factorization： the first two layers of signals are detected by ML algorithm, so the performance of the first detection level is improved. The rest layers of signals are detected by QR factorization algorithm. Theoretical analysis and experimental results show that the performance of Sorted ML-QR algorithm is superior to the conven- tional layered space-time detection algorithms, and the performance improvement is more obvious when the SNR is larger.

关 键 词：垂直分层空时码 MIMO 排序ML—QR QR分解 最大似然检测 

分 类 号：TN911.23[电子电信—通信与信息系统]