减小2×2 MIMO系统最大似然检测的复杂度  

Complexity Reduction of Maximum Likelihood Detection of 2 × 2 MIMO System

作　　者：胡东伟[1,2] 

机构地区：[1]中国电子科技集团公司第五十四研究所,河北石家庄050081 [2]通信网信息传输与分发技术重点实验室,河北石家庄050081

出　　处：《无线电工程》2014年第10期22-26,共5页Radio Engineering

基　　金：国家部委基金资助项目

摘　　要：2×2多输入多输出(Multiple Input Multiple Output,MIMO)系统在第4代移动通信、无线局域网等场合广泛应用。分析了现有的2×2 MIMO系统的检测算法及其复杂度。针对目前的最大似然算法复杂度过高的问题,提出了3种降低复杂度的方法:预三角化方法、以曼哈顿距离取代欧氏距离的方法和软信息的近似提取方法。其中前2种方法能保持最大似然的检测性能,第3种方法以2.5 dB的性能损失(误码率10-3时)为代价,将复杂度降低一半。The 2×2 MIMO systems are widely used in the fourth generation of mobile communication systems and Wireless Local Area Networks （WLAN）.The existing detection techniques and their complexity for 2×2 MIMO systems are firstly analyzed.In order to reduce the complexity of the Maximum Likelihood （ML） detection of 2×2 MIMO systems,three methods are proposed：firstly,triangu-larization is performed before the detection;secondly,Euclidean distance is replaced by Manhattan distance;thirdly,an approximate soft information extracting algorithm is proposed.With the first two techniques,ML performance is held by the first two methods.The third method reduces the complexity by a half with the cost of 2.5 dB performance degradation when the Bit Error Rate （ BER） is 10^-3 .

关 键 词：多输入多输出系统 最大斯然检测 软信息 

分 类 号：TN929[电子电信—通信与信息系统]