| 一种基于时频分析的盲信号分离算法 |
| |
| 引用本文： | 何文雪,王林,谢剑英.一种基于时频分析的盲信号分离算法[J].上海交通大学学报,2005,39(12):2061-2065. |
| |
| 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,电子信息与电气工程学院,上海,200030 |
| |
| 基金项目： | 青岛大学博士科研基金（06300353）资助项目 |
| |
| 摘 要： | 利用模糊函数的时频分布特征选取时频点，提出一种新的时频域盲信号分离算法．与传统的维纳时频分布相比，不仅显著缩小了时频点的选择范围，而且不需要设定阈值．算法利用一种非正交联合对角化方法求取分离矩阵，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．仿真试验表明，该算法计算量小，分离精度高，可实现高斯源或相关源信号的盲分离．
|
| 关 键 词： | 盲分离 时频分析 模糊函数 联合对角化 |
| 文章编号： | 1006-2467(2005)12-2061-05 |
| 收稿时间： | 2004-11-10 |
| 修稿时间： | 2004年11月10 |
A Blind Source Separation Algorithm by Using Time-Frequency Signal Analysis |
| |
| HE Wen-xue,WANG Lin,XIE Jian-ying.A Blind Source Separation Algorithm by Using Time-Frequency Signal Analysis[J].Journal of Shanghai Jiaotong University,2005,39(12):2061-2065. |
| |
| Authors: | HE Wen-xue WANG Lin XIE Jian-ying |
| |
| Institution: | 1. College of Automation Eng. , Qingdao Univ. , Qingdao 266071, China; 2. School of Electronic, Information and Electrical Eng. , Shanghai Jiaotong Univ. , Shanghai 200030 |
| |
| Abstract: | The new blind source separation(BSS) algorithm uses the time-frequency distribution feature of ambiguity function to select the proper time-frequency point,which decreases the selection area greatly compared with the conventional Wigner distribution and no threshold is needed.A non-orthogonal joint diagonalization method is used here to get the separation matrix,which adapts to the situation where the source signals are gaussians or spatial correlated.Simulations were made to validate the effectiveness of the algorithm. |
| |
| Keywords: | blind source separation time-frequency analysis ambiguity function joint diagonalization |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
