A penalized mutual information criterion for blind separation of convolutive mixtures

Rhabi, Mohammed El; Gellé, Guillaume; Fenniri, H.; Delaunay, G.

doi:10.1016/j.sigpro.2004.06.015

Cited by 17 publications

(14 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The mixtures were separated using mutual information (MI) method shown in ( [10,11]), MI method where the data are pre-whitening (MI BL) and our proposed algorithm (MI TV).…”

Section: Numerical Resultsmentioning

confidence: 99%

“…To make it easier to understand, let us consider now a bidimensional random vector y = (y 1 (n), y 2 (n)) T . The independence of the components y 1 (n) and y 2 (n ) is needed for all n and n to ensure the separation, in a different way the independence of y 1 (n) and y 2 (n − m), for all n and at all lags m. As in [10,11], we define the separating criterion J sep by:…”

Section: The Simultaneous Bss-denoising Proceduresmentioning

confidence: 99%

See 1 more Smart Citation

On convolutive Blind Source Separation in a noisy context and a total variation regularization

Boulmezaoud¹,

Rhabi²,

Fenniri³

et al. 2010

2010 IEEE 11th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC)

View full text Add to dashboard Cite

We propose a new strategy for improving classical Blind Source Separation (BSS) methods. This strategy consists in denoising both the observed and the estimated source signals, and is based on the minimization of a regularized criterion which takes into account the Total Variation of the signal. We prove by the way that the method leads to a projection problem which is solved by means of projected gradient algorithm. The effectiveness and the robustness of the proposed separating process are shown on numerical examples.

show abstract

“…The mixtures were separated using mutual information (MI) method shown in ( [10,11]), MI method where the data are pre-whitening (MI BL) and our proposed algorithm (MI TV).…”

Section: Numerical Resultsmentioning

confidence: 99%

Section: The Simultaneous Bss-denoising Proceduresmentioning

confidence: 99%

On convolutive Blind Source Separation in a noisy context and a total variation regularization

Boulmezaoud¹,

Rhabi²,

Fenniri³

et al. 2010

2010 IEEE 11th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC)

View full text Add to dashboard Cite

show abstract

“…The main drawback of this method is that the convergence of the algorithm is affected by the normalization process. In order to overcome the normalization constraint, a penalization term [7] is added to (4) as…”

Section: The Penalized Mutual Information Criterionmentioning

confidence: 99%

“…Hence, a suitable regularization is often implemented to overcome nonunique solutions [19]. Due to the practicability and the relative uniqueness of the solution, the blind separation of post-nonlinear (PNL) mixtures plays an important role in nonlinear ICA [21], [9], [15], [11], [20], [2], [3], [7], [13], where the mixtures consist of a matrix and component-wise invertible nonlinearities. Taleb and Jutten [15] use a multilayer perception (MLP) to approximate the inverse nonlinear function and give a versatile learning procedure based on the minimization of mutual information (MMI).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Using the Penalized Mutual Information Criterion in the Multivariate Edgeworth-Expanded Gaussian Mixture Density for Blind Separation of Convolutive Post-Nonlinear Mixtures

Chen

Shi

2007

Circuits Syst Signal Process

View full text Add to dashboard Cite

This paper proposes the blind separation of convolutive post-nonlinear (CPNL) mixtures based on the minimization of the penalized mutual information criterion. The proposed algorithm is based on the estimation score function difference (SFD) and the Newton optimization. Compared with the blind source separation of a linear mixture, the separation performance of a nonlinear mixture is strongly related to the accuracy of the score function estimation. Under this framework, the multivariate Edgeworth-expanded Gaussian mixture density is adopted to estimate the SFD, which preserves the higher-order statistical structure of the data as compared to the nonparametric density estimation. Also, the Newton optimization converges faster than the steepest descent gradient. In order to calculate the Hessian matrix, the Taylor expansion of the penalized mutual information criterion is extended to second order. The minimization of the penalized mutual information criterion ensures a priori normalization of the estimated sources, thus avoiding scale indeterminacy. The proposed algorithm has a better performance, and at the same time it speeds up the convergence. Simulations with computer-generated data and synthetic realworld data show the effectiveness of the proposed algorithm.

show abstract

Separability of Convolutive Mixtures: Application to the Separation of Combustion Noise and Piston-Slap in Diesel Engine

Akil

Servière

2006

Independent Component Analysis and Blind Signal Separation

View full text Add to dashboard Cite

A penalized mutual information criterion for blind separation of convolutive mixtures

Cited by 17 publications

References 3 publications

On convolutive Blind Source Separation in a noisy context and a total variation regularization

On convolutive Blind Source Separation in a noisy context and a total variation regularization

Using the Penalized Mutual Information Criterion in the Multivariate Edgeworth-Expanded Gaussian Mixture Density for Blind Separation of Convolutive Post-Nonlinear Mixtures

Separability of Convolutive Mixtures: Application to the Separation of Combustion Noise and Piston-Slap in Diesel Engine

Contact Info

Product

Resources

About