| 基于模糊函数的单步盲信号分离算法 |
| |
| 引用本文： | 何文雪,王林,谢剑英.基于模糊函数的单步盲信号分离算法[J].系统仿真学报,2005,17(5):1261-1264. |
| |
| 作者姓名： | 何文雪 王林 谢剑英 |
| |
| 作者单位： | 1. 青岛大学自动化工程学院,青岛,266071
2. 上海交通大学自动化所,上海,200030 |
| |
| 基金项目： | 国家863基金项目(2001AA422420-02) |
| |
| 摘 要： | 基于模糊函数的分布特性提出了一种新的时频盲信号分离算法，利用信号模糊函数分布的对称性与居中性选取相应的时频点进行分析，与维纳时频分布相比可减小工作量，而且不需要设定闽值。算法利用一种非正交联合对角化方法求取分离矩阵，不需对观测信号进行预白化处理。仿真表明在原始信号具有互相关的情况下也能达到良好的分离性能。
|
| 关 键 词： | 盲分离 时频分析 模糊函数 联合对角化 |
| 文章编号： | 1004-731X(2005)05-1261-04 |
| 修稿时间： | 2004年5月11日 |
Single Step Blind Source Separation Algorithm using Ambiguity Function |
| |
| HE Wen-xue,WANG Lin,XIE Jian-ying.Single Step Blind Source Separation Algorithm using Ambiguity Function[J].Journal of System Simulation,2005,17(5):1261-1264. |
| |
| Authors: | HE Wen-xue WANG Lin XIE Jian-ying |
| |
| Abstract: | A new time-frequency domain blind source separation algorithm by using ambiguity function is presented in this paper. The algorithm uses the symmetric distribution and concentrated distribution properties of the ambiguity function to select the corresponding analysis points, which decreases the computing cost comparing with Winner time-frequency analysis, and no threshold is needed. A non-orthogonal joint diagonal method is utilized here in order to remove the pre-whitening process. Simulations have shown that good separation performance could be obtained even when the source signals are temporarily correlated. |
| |
| Keywords: | blind source separation time-frequency analysis ambiguity function joint diagonal |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
