Blind Identification Of Independent Components With Higher-order Statistics
- 24 August 2005
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 157-162
- https://doi.org/10.1109/hosa.1989.735288
Abstract
We consider a N-variate random process consisting in N (or less) independent components corrupted by additive noise. Identifying (or separating) these components without any a-priori information about their structure (blind identification) is theorically possible under the independence assumption. We propose a blind algorithm based on second- and fourth-order statistics. Its main feature is the use of tensor formalism allowing blind identification to be performed as a generalized eigen-decomposition of the "quadricovariance tensor". All non-Gaussian components (and possibly one Gaussian component) are separated even if they have identical probability distributions. Simulations in Array Processing context are included.Keywords
This publication has 7 references indexed in Scilit:
- Blind Identification Of Independent Components With Higher-order StatisticsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- Cumulant-based approach to the harmonic retrieval problemPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Sources indentification: a solution based on the cumulantsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Harmonic decomposition methods in cumulant domains (array processing)Published by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Identification of linear stochastic systems via second- and fourth-order cumulant matchingIEEE Transactions on Information Theory, 1987
- Time Series: Data Analysis and Theory.Published by JSTOR ,1981
- Robust identification of a nonminimum phase system: Blind adjustment of a linear equalizer in data communicationsIEEE Transactions on Automatic Control, 1980