Blind Source Separation in Post-Nonlinear Mixtures Using Competitive Learning, Simulated Annealing, and a Genetic Algorithm

Rojas, Fernando; Puntonet, Carlos G.; Rodríguez-Álvarez, Manuel; Rojas, Ignacio; Martı́n-Clemente, Rubén

doi:10.1109/tsmcc.2004.833297

Cited by 42 publications

(27 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The nonlinear ICA algorithm, proposed in (Rojas et al, 2004), was applied to given training 7-channel EEG data sets associated to the hand movement imagination and resting state.…”

Section: Resultsmentioning

confidence: 99%

“…In (Rojas et al, 2004), it has been added some extra constraints to the nonlinear mixture so that the nonlinearities are independently applied in each channel after a linear mixture. As figure 3 shows, the proposed algorithm in (Rojas et al, 2004) needs to estimate two different mixtures: a family of nonlinearities g which approximates the inverse of the nonlinear mixtures f and a linear unmixing matrix W which approximates the inverse of the linear mixture A. For the demixing system, first we need to approximate i g , which is the inverse of the nonlinear function in each channel, and then separate the linear mixing by applying W to the output of the i g nonlinear function:…”

Section: Nonlinear Icamentioning

confidence: 99%

“…By using relations (12) and (13) The parameter vector j p s h o u l d b e d e t e r m i n e d s o t h at the inverse of the mutual information of the output sources i y is maximized. To achieve this objective, can be defined the following criterion (Rojas et al, 2004):…”

Section: Nonlinear Icamentioning

confidence: 99%

See 2 more Smart Citations

Nonlinear Independent Component Analysis for EEG-Based Brain-Computer Interface Systems

Oveisi¹,

Oveisi²,

Efranian³

et al. 2012

Independent Component Analysis for Audio and Biosignal Applications

View full text Add to dashboard Cite

“…The nonlinear ICA algorithm, proposed in (Rojas et al, 2004), was applied to given training 7-channel EEG data sets associated to the hand movement imagination and resting state.…”

Section: Resultsmentioning

confidence: 99%

Section: Nonlinear Icamentioning

confidence: 99%

See 1 more Smart Citation

Nonlinear Independent Component Analysis for EEG-Based Brain-Computer Interface Systems

Oveisi¹,

Oveisi²,

Efranian³

et al. 2012

Independent Component Analysis for Audio and Biosignal Applications

View full text Add to dashboard Cite

“…They are also easily parallelizable and their evaluation function can be any that assigns to each individual a real value into a partially ordered set (poset). GAs have already been successfully applied to linear and post-nonlinear blind source separation [12].…”

Section: Justifying the Use Of A Ga For Blind Deconvolutionmentioning

confidence: 99%

A Canonical Genetic Algorithm for Blind Inversion of Linear Channels

Rojas

Solé-Casals

Monte-Moreno

et al. 2006

Independent Component Analysis and Blind Signal Separation

View full text Add to dashboard Cite

Abstract.It is well known the relationship between source separation and blind deconvolution: If a filtered version of an unknown i.i.d. signal is observed, temporal independence between samples can be used to retrieve the original signal, in the same manner as spatial independence is used for source separation. In this paper we propose the use of a Genetic Algorithm (GA) to blindly invert linear channels. The use of GA is justified in the case of small number of samples, where other gradient-like methods fails because of poor estimation of statistics.

show abstract

“…Tan and Wang in [14] used GA to solve the nonlinear BSS problem using higher order statistics where the sources have been estimated regardless of the indeterminacies of permutation and scaling. Rajas et al in [15] applied the GA based method for signal separation from their post nonlinear mixtures. GA has been directly applied to ICA problem for first time in [16] for denoising the electrocardiogram (ECG) signals where the method estimates only one independent component i.e.…”

Section: Introductionmentioning

confidence: 99%

Constrained genetic algorithm based independent component analysis

Acharya

Panda

Lakshmi

2007

2007 IEEE Congress on Evolutionary Computation

View full text Add to dashboard Cite

Abstract-Independent Component Analysis, a computationally efficient statistical signal processing technique, has been an area of interest for researchers for many practical applications in various fields of science and engineering. The present paper proposes a constrained genetic algorithm optimization based independent component analysis assuming a noise free Independent Component Analysis (ICA) model. It investigates on the application and performance of the popular evolutionary computation technique GA in independent component analysis problem. It is observed that the proposed constrained genetic algorithm optimization based ICA overcomes the long standing permutation ambiguity and recovers the independent components in a fixed order which is dependent on the statistical characteristics of the signals to be estimated. The constrained GA based ICA has also been compared with the most popular fast ICA algorithm.

show abstract

Blind Source Separation in Post-Nonlinear Mixtures Using Competitive Learning, Simulated Annealing, and a Genetic Algorithm

Cited by 42 publications

References 18 publications

Nonlinear Independent Component Analysis for EEG-Based Brain-Computer Interface Systems

Nonlinear Independent Component Analysis for EEG-Based Brain-Computer Interface Systems

A Canonical Genetic Algorithm for Blind Inversion of Linear Channels

Constrained genetic algorithm based independent component analysis

Contact Info

Product

Resources

About