基于源信号数目估计的欠定盲分离

谭北海; 谢胜利

doi:10.3724/SP.J.1146.2006.01414

基于源信号数目估计的欠定盲分离

doi: 10.3724/SP.J.1146.2006.01414

基金项目:

国家杰出青年自然科学基金(60325310)，广东省自然科学团队研究项目(04205783)，广东省自然科学基金(05103553,05006508)，国家自然科学基金重点项目(U0635001)和科技部重大基础前期研究专项(2005CCA04100)资助课题

计量
- 文章访问数: 3138
- HTML全文浏览量: 131
- PDF下载量: 1004
- 被引次数: 0
出版历程
- 收稿日期: 2006-09-18
- 修回日期: 2007-03-19
- 刊出日期: 2008-04-19

Underdetermlned Blind Separation Based on Source Signals Number Estimation

摘要

摘要: 该文利用欠定盲分离下稀疏源信号的特点，估计源信号的数目且恢复源信号。通常在用两步法来解决欠定盲分离时，首先利用K-均值算法对观测信号聚类估计出混叠矩阵，最后利用最短路径法来恢复源信号，但是在以往的算法中，第1步估计混叠矩阵时，通常假设源信号数目是已知的，从而进行K-均值聚类，而事实上源信号数目根本无法知道，因此对源信号数目的估计对两步法有很重要的影响。因此本文提出了一种新的两步法算法，其中第1步利用稀疏源信号反映在观测信号中的特征来准确地估计出稀疏源信号的数目，且能得到混叠矩阵，从而恢复源信号。最后的仿真结果，以及与通常的K-均值聚类算法对比的仿真结果说明了此算法的可行性和优异的性能。
- 信号处理; 稀疏表示; 欠定盲分离; 混叠矩阵; 两步法
Abstract: This paper gives a new method to estimate the number of source signals and recover them by the characteristics of sparse source signals in underdetermined blind separation. It is well known that source signals can be recovered through the two-step algorithms generally. The first step is to estimate the mixture matrix by K-means clustering algorithm using the sensor signals, and then, the shortest path algorithm is used to recover source signals, whereas, people suppose that the number of source signals is known when they estimate the mixture matrix by the K-means clustering algorithm generally. In fact, the number of source signals is unknown or blind, so it is very important to estimate the number of source signals. In this paper, a new two-step algorithm is proposed, which not only can estimate the number of source signals but also get the mixture matrix instead of K-means algorithm through the characteristics of sensor signals. The last simulation results show the algorithm simply, efficient and good performance.

HTML全文

参考文献(1)

Jutten C and Herault J. Blind separation of sources, Part I: Anadaptive algorithm based on neuromimetic[J].Signal Processing.1991, 24(1):1-10[2]Hyvarinen A and Oja E. Independent component analysis:algorithms and applications[J].Neural Networks.2000, 13(4-5):411-430[3]Xie Shengli, He Zhaoshui, and Fu Yuli. A note on Stonesconjecture of blind separation[J].Neural Computation.2004, 17(2):245-319[4]Hyvarinen A. Fast and robust fixed-point algorithms forindependent component analysis. IEEE Trans. on NeuralNetworks, 1999, 10(3): 626-634.[5]Zhang Jinlong, Xie Shengli, and He Zhaoshui. Separability theoryfor blind signal separation. Acta. Automatica Sinica, 2004, 30(3):337-344.[6]Belouchrani A and Cardoso J F. Maximum likelihood sourceseparation for discrete sources. In Proceeding of DUSIPCO,Edinburg, Scotland, 1994, 768-771.[7]Zibulevsky M and Pearlmutter B A. Blind source separation bysparse decomposition in a signal dictionary[J].Neural computation.2001, 13(4):863-882[8]Li Yuanqing, Cichocki A, and Amari S. Analysis of sparserepresentation and blind source separation[J].Neural Computation.2004, 16(6):1193-1234[9]Wax M and Kailath T. Detection of signals by informationtheoretic criteria. IEEE Trans. on ASSP, 1985, 33(2): 276-280.[10]张洪渊, 贾鹏, 史习智. 确定盲分离中未知信号源个数的奇异值分解法. 上海交通大学学报, 2001, 35(8): 1155-1158.[11]章新华, 张安清, 孙剑平. 一种信号源数目的盲估计方法. 系统工程与电子技术, 2001, 23(9): 9-11.[12]张翼, 柯亨玉, 文必洋, 吴雄斌. 相位法估计信号源数. 武汉大学学报, 2003, 49(1): 137-140.[13]李广彪, 许士敏. 基于源数估计的盲源分离. 系统仿真学报, 2006,18(2): 485-488.[14]Lee T W, Lewicki M S, Girolami M, and Sejnowski T J. Blindsource separation of more sources than mixtures usingovercomplete representation[J].IEEE Signal Processing Letter.1999,6(4):87-90[15]Lewicki M S and Sejnowski T J. Learning overcompleterepresentations, Neural Computation, 2000, 12(1): 337-365.[16]Bofill P and Zibulevsky M. Underdetermined source separationusing sparse representation[J].Signal Processing.2001, 81(11):2353-2362[17]李远清, 张丽清. 一种数字图像盲增强的新算法. 控制理论与应用, 2003, 20(4): 525-528.

施引文献

资源附件(0)

访问统计