一种改进的递推最小二乘盲多用户检测算法  

Improved Blind Multi-User Detection Algorithm Based on Recursive Least Square Criterion

作　　者：曾召华[1] 刘贵忠[1] 赵建平[2] 

机构地区：[1]西安交通大学电子与信息工程学院,西安710049 [2]中兴通讯股份有限公司西安研究所

出　　处：《西安交通大学学报》2003年第6期631-634,共4页Journal of Xi'an Jiaotong University

基　　金：陕西省教育厅自然科学专项基金资助项目 (0 1 jk1 94).

摘　　要：为提高基于递推最小二乘(RLS)算法的盲多址干扰抑制检测器的稳态输出性能,根据空间正交投影算法和自适应判决反馈控制算法,提出一种基于RLS算法的自适应判决反馈盲干扰抑制检测器,该检测器在启动阶段,等价为一种基于RLS准则的盲干扰抑制检测器,在稳态收敛阶段,等价为一种有训练数据的RLS干扰抑制检测器.既保持了最小输出能量(MOE)检测器的全局收敛性,又具有最小均方误差检测器高输出信干比的优点.计算机仿真结果表明,该检测器性能优于已有的基于MOE准则的检测器,且能应用于时变信道.Based on recursive-least-square (RLS) algorithm using orthogonal projection technology and adaptive decision feed back technology, a blind interference suppression detector is proposed to improve performance of RLS-MOE detector. During startup, the detector is a blind detector based on RLS criterion essentially. During convergence, it is a detector with known data symbols based on RLS criterion, which achieves the superior convergence performance of the MOE blind detector and the high output signal-interference rate of the minimum mean square error (MMSE) detector. Computer simulation results show that the convergence performance of the proposed detector is better than those of blind detectors based on least-mean-square (LMS) and RLS criterion, and it can be used in time-varying channel.

关 键 词：自适应判决反馈 递推最小二乘 盲多用户检测 最小输出能量 

分 类 号：TN929.5[电子电信—通信与信息系统]