Topographic map formation of factorized Edgeworth-expanded kernels
Neural Networks - 2006 Special issue: Advances in self-organizing maps--WSOM'05
On the entropy minimization of a linear mixture of variables for source separation
Signal Processing - Special issue: Information theoretic signal processing
Separation of statistically dependent sources using an L2-distance non-Gaussianity measure
Signal Processing - Special section: Distributed source coding
Blind spectral unmixing by local maximization of non-Gaussianity
Signal Processing
Analysis of the Kurtosis-Sum Objective Function for ICA
ISNN '08 Proceedings of the 5th international symposium on Neural Networks: Advances in Neural Networks
A general procedure for learning mixtures of independent component analyzers
Pattern Recognition
A source adaptive independent component analysis algorithm through solving the estimating equation
Expert Systems with Applications: An International Journal
Blind separation of piecewise stationary non-Gaussian sources
Signal Processing
Fast kernel-based independent component analysis
IEEE Transactions on Signal Processing
Robust independent component analysis using quadratic negentropy
ICA'07 Proceedings of the 7th international conference on Independent component analysis and signal separation
Complex independent component analysis by entropy bound minimization
IEEE Transactions on Circuits and Systems Part I: Regular Papers
Independent component analysis by entropy bound minimization
IEEE Transactions on Signal Processing
ICA mixtures applied to ultrasonic nondestructive classification of archaeological ceramics
EURASIP Journal on Advances in Signal Processing - Special issue on signal processing in advanced nondestructive materials inspection
Sequential extraction algorithm for BSS without error accumulation
ISNN'05 Proceedings of the Second international conference on Advances in neural networks - Volume Part II
A new image protection and authentication technique based on ICA
ISNN'05 Proceedings of the Second international conference on Advances in Neural Networks - Volume Part I
Non-parametric ICA algorithm for hybrid sources based on GKNN estimation
CIS'05 Proceedings of the 2005 international conference on Computational Intelligence and Security - Volume Part II
Fast kernel density independent component analysis
ICA'06 Proceedings of the 6th international conference on Independent Component Analysis and Blind Signal Separation
Multi-level independent component analysis
ISNN'06 Proceedings of the Third international conference on Advances in Neural Networks - Volume Part I
Unified parametric and non-parametric ICA algorithm for arbitrary sources
ISNN'06 Proceedings of the Third international conference on Advances in Neural Networks - Volume Part I
A novel kurtosis-dependent parameterized independent component analysis algorithm
ISNN'06 Proceedings of the Third international conference on Advances in Neural Networks - Volume Part I
An efficient score function generation algorithm with information maximization
ICNC'05 Proceedings of the First international conference on Advances in Natural Computation - Volume Part I
Newton-like methods for nonparametric independent component analysis
ICONIP'06 Proceedings of the 13 international conference on Neural Information Processing - Volume Part I
Exterior penalty function method based ICA algorithm for hybrid sources using GKNN estimation
ICONIP'06 Proceedings of the 13 international conference on Neural Information Processing - Volume Part I
A semiparametric approach to source separation using independent component analysis
Computational Statistics & Data Analysis
| Hi-index | 0.01 |
In this paper, we introduce a novel independent component analysis (ICA) algorithm, which is truly blind to the particular underlying distribution of the mixed signals. Using a nonparametric kernel density estimation technique, the algorithm performs simultaneously the estimation of the unknown probability density functions of the source signals and the estimation of the unmixing matrix. Following the proposed approach, the blind signal separation framework can be posed as a nonlinear optimization problem, where a closed form expression of the cost function is available, and only the elements of the unmixing matrix appear as unknowns. We conducted a series of Monte Carlo simulations, involving linear mixtures of various source signals with different statistical characteristics and sample sizes. The new algorithm not only consistently outperformed all state-of-the-art ICA methods, but also demonstrated the following properties: 1) Only a flexible model, capable of learning the source statistics, can consistently achieve an accurate separation of all the mixed signals. 2) Adopting a suitably designed optimization framework, it is possible to derive a flexible ICA algorithm that matches the stability and convergence properties of conventional algorithms. 3) A nonparametric approach does not necessarily require large sample sizes in order to outperform methods with fixed or partially adaptive contrast functions.