期刊文献+

一种加权增强的欠定盲辨识方法

A weighted enhauced underdetermined mixtures separation method
在线阅读 下载PDF
导出
摘要 针对欠定盲源分离混合矩阵问题,提出了一种基于二阶统计量平行因子分解,加权增强最小二乘法的欠定混合盲辨识方法。该算法不需要源信号满足稀疏性要求,仅在源信号满足相互独立和最多一个高斯信号的条件下,将独立源信号的空间协方差矩阵构建三阶张量,采用加权增强最小二乘法实现张量的标准分解,完成混合矩阵的估计。由于平行因子分解的唯一性在欠定条件下依然成立,该算法可以解决欠定盲源分离问题。仿真实验结果表明:提出的算法在计算欠定混合时具有很好的辨识效果,而且实现简单,可满足实际应用的要求。 To solve the problem of the underdetermined blind source separation, this paper proposes a algorithm for btind identification of underdetermined mixtures based on parallel factor decomposition of covariance matrix of the observed signal and weight enhanced alternating least squares, without the need of the sources are quite sparse. Because the parallel factor decomposition still satisfied unique identifiability in underdetermined situation, the proposed algorithm can solve the underdetermined blind source separation problem successfully. Simulation results demonstrate that the performance and effectiveness of the proposed algorithm is very better for underdetermined mixed. The algorithm is relatively simple, which can satisfy the demand of engineering application.
出处 《电子技术应用》 北大核心 2015年第3期101-104,110,共5页 Application of Electronic Technique
基金 山西省自然科学基金项目(2013011016-1) 教育部博士学科点专项科研基金项目(2011081047) 山西省自然科学青年基金项目(2013021016-1)
关键词 欠定混合 盲辨识 增强线搜索 最小二乘法 underdetermined mixtures blind identification enhanced line search least squares
作者简介 李振涛(1987-),女,在渎硕士,主要研究方向:盲信号处理。 李鸿燕(1973-),通信作者,女,博士研究生,副教授,主要研究方向:信号处理、模式识别,E-mail:869333623@qq.com。
  • 相关文献

参考文献8

  • 1LIEVEN De LATHAUWER.Blind identification of underde- termined mixtures by simultaneous matrix diagonalization[J]. IEEE Transactions on Signal Processing, 2008,56(3): 1096-1105.
  • 2PETR Tichavsky,ZBYNEK Koldovsky.Weight adjusted tensor method for blind separation of underdetermined mixtures of nonstationary sources[J].IEEE Transactions on Signal Processing, 2011 , 59 : 1037-1047.
  • 3KRUSKAL J B.Three-way arrays:Rank and uniqueness of trilinear decompositions, with application to arithmetic complexity and statistics, Linear Algebra Appl., 1977(18): 95-138.
  • 4STEGEMAN A,SIDIROPOULOS N D.On kruskal's uniqueness condition for the Candecomp/ PARAFAC decomposition[C].Linear Algebra and its Applications,2007 (420) 540-552.
  • 5NICHOLAS D.SIDIROPOULOS, RASMUSBRO.On the Uni- queness of Multilinear Decomposition of N-way Arrays[J]. J. Chemometrics, 2000,14 (3) : 229- 239.
  • 6RAJIH M ,COMON P ,HARSI-IMAN R A.Enhanced line search :A novel method to accelerate PARAFAC[J].SIAM Journal on Matrix Analysis, 2008,30 (3) : 1148 - 1171.
  • 7NION D ,DE LATHAUWER L.Line Search computation of the Block Factor Model for blind muhi-user access in wireless communications[C].Advances in Wireless Communi- cations(SPAWC) France,Cannes,.IEEE Workshop on Signal Processing, 2006.
  • 8NION D,DE LATHAUWER I.An enhanced line Search scheme for complex-valued tensor decompositions.Application in DS-CDMA[J].Signal Processing, 2008,3(88) : 749-755.

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部