期刊文献+

基于源信号统计独立性的ICA方法的不确定性研究 被引量:6

Study on the indeterminacy of the statistical independence based ICA methods
下载PDF
导出
摘要 3种基于源信号统计独立性的ICA方法———极大似然法、最大信息法和最小互信息法是等价的。从这3种方法的计算公式出发,分析了自然梯度算法的收敛条件,指出了ICA问题解的不确定性和近似性的根源。通过论证表明,在源信号都属于指数型的前提下,为亚高斯型和超高斯型源信号适当选择的作为评价函数的非线性函数具有很好的韧性。 The convergence condition for the natural gradient algorithm is analyzed based on the calculation equation of the maximum likelihood, the informax and the minimum mutual information ICA algorithms. The reasons of the indeterminacy and the approximation property on the solution of ICA are presented. If all sources are exponential, the property-related nonlinear evaluation functions for both super-and sub-Gaussian sources are robust. Simulation results show the indeterminacy of these algorithms.
出处 《系统工程与电子技术》 EI CSCD 北大核心 2004年第4期556-559,共4页 Systems Engineering and Electronics
基金 国家自然科学基金(30170259 60172072) 国家"973"专项基金(2001CCA00700) 辽宁省科学技术基金(2001101057)资助课题
关键词 盲源分离 独立分量分析 不确定性 极大似然法 blind source separation ICA indeterminate property maximum likelihood
  • 相关文献

参考文献11

  • 1Comon P. Independent Component Analysis: A New Concept[J]. Signal Processing,1994,34:287-314.
  • 2Hyvarinen A,Oja E. Independent Component Analysis: Algorithms and Applications[J]. Neural Networks,2000,13:411-430.
  • 3Hyvarinen A. Survey on Independent Component Analysis[J]. Neural Computing Surveys,1999,2:94-128.
  • 4Hyvarinen A,Pajunen P. Nonlinear Independent Component Analysis: Existence and Uniqueness Results[J]. Neural Networks,1999,12:429-439.
  • 5Lee T,Sejnowski T J. Independent Component Analysis for Sub-Gaussian and Super-Gaussian Mixtures[C]. In 4th Joint Symposium on Neural Computation,Institute for Neural Computation,1997,7:132-139.
  • 6Bell A J,Sejnowski T J. An Information-Maximization Approach to Blind Separation and Blind Deconvolution[J]. Neural Computation,1995,7(6):1129-1159.
  • 7Lee T,Girolami M,Bell A J, et al. A Unifying Information Theoretic Framework for Independent Component Analysis[J]. Computers and Mathematics with Applications,2000,39:1-21.
  • 8Cardoso J F. Informax and Maximum Likelihood for Blind Source Separation[J]. IEEE Signal Processing Letters,1997,4:112-114.
  • 9Hyvarinen A. Independent Component Analysis in the Presence of Gaussian Noise by Maximizing Joint Likelihood[J]. Neurocomputing,1998,22:49-67.
  • 10Hyvariene A,Juha Karhunen,Erkki Oja. Independent Component Analysis[M]. New York:John Wiley & Sons Inc.,2001.

二级参考文献10

  • 1Hyvarinen A.Fast and robust fixed-point algorithm for independent component analysis[].IEEE Transactions on Neural Networks.1999
  • 2Amari SI,Cichocki AC.Adaptive blind signal processing-Neural network approaches[].Proceedings of the IEEE.1998
  • 3Cardoso JF.Blind signal processing[].Proceedings of the IEEE.1998
  • 4Cardoso JF.Higher order contrasts for independent component analysis[].Neural Computation.1999
  • 5Amari SI.Natural gradient works effciently in learning[].Neural Computation.1998
  • 6Cichocki A,Unbehanen R,Rummert E.Robust learning algorithm for blind separation of sources[].Electronics Letters.1994
  • 7Lee TW,Amari SI,Cichocki AC.Independent component analysis using an extended infomax algorithm for mixed sub-Gaussian and super -Gaussian sources[].Neural Computation.1999
  • 8Cardoso JF,Laheld BH.Equivariant adaptive source separation[].IEEE Transactions on Signal Processing.1996
  • 9Comon P.Independent component analysis:A new concept ?[].Signal Processing.1994
  • 10Hyvarinen A,Oja E.A fast fixed-point algorithm for independent component analysis[].Neural Computation.1997

共引文献57

同被引文献57

引证文献6

二级引证文献23

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