Abstract:Independent component analysis or blind source separation extracts independent signals from their linear mixtures without assuming prior knowledge of their mixing coefficients. It is known that the independent signals in the observed mixtures can be successfully extracted except for their order and scales. In order to resolve the indeterminacy of scales, most learning algorithms impose some constraints on the magnitudes of the recovered signals. However, when the source signals are nonstationary and their average magnitudes change rapidly, the constraints force a rapid change in the magnitude of the separating matrix. This is the case with most applications (e.g., speech sounds, electroencephalogram signals). It is known that this causes numerical instability in some cases. In order to resolve this difficulty, this article introduces new nonholonomic constraints in the learning algorithm. This is motivated by the geometrical consideration that the directions of change in the separating matrix should be orthogonal to the equivalence class of separating matrices due to the scaling indeterminacy. These constraints are proved to be nonholonomic, so that the proposed algorithm is able to adapt to rapid or intermittent changes in the magnitudes of the source signals. The proposed algorithm works well even when the number of the sources is overestimated, whereas the existent algorithms do not (assuming the sensor noise is negligibly small), because they amplify the null components not included in the sources. Computer simulations confirm this desirable property.

Nonlinear blind source separation exploiting spatial nonstationarity

Modelling multivariate spatio-temporal data with identifiable variational autoencoders

Blind nonlinear hyperspectral unmixing based on constrained kernel nonnegative matrix factorization

On Cokriging, Neural Networks, and Spatial Blind Source Separation for Multivariate Spatial Prediction

Nonlinear adaptive blind source separation based on kernel function

Separation of Uncorrelated Stationary Time Series Using Autocovariance Matrices

Blind source separation with nonlinear autocorrelation and non-Gaussianity

Nonholonomic Orthogonal Learning Algorithms for Blind Source Separation

An Adaptive Approach to Blind Source Separation Using a Self-Organzing Map and a Neural Gas

Parameter-adaptive variational autoencoder for linear/nonlinear blind source separation

Visual Parameter Selection for Spatial Blind Source Separation

Nonlinear signals separation of adaptive natural gradient learning

A novel approach for underdetermined blind sources separation in frequency domain

Blind Source Separation with Time Series Variational Bayes Expectation Maximization Algorithm.

Efficient blind nonparametric dependent signal extraction algorithm for determined and underdetermined mixtures.

Neural Network Approaches to Nonlinear Blind Source Separation.

Blind Nonparametric Determined and Underdetermined Signal Extraction Algorithm for Dependent Source Mixtures

Blind Source Separation From Hybrid Mixture Based On Nonlinear Infomax Approach

Blind Source Separation in Post-nonlinear Mixtures Using Natural Gradient Descent and Particle Swarm Optimization Algorithm

Underdetermined Blind Source Separation Using Sparse Coding

Blind Inversion of Wiener System for Single Source Using Nonlinear Blind Source Separation