Blind separation of multichannel electrogastrograms using independent component analysis based on a neural network

The instantaneous noise-free linear mixing model in independent component analysis is largely a solved problem under the usual assumption of independent nongaussian sources and full column rank mixing matrix. However, with some prior information on the sources, like positivity, new analysis and perhaps simplified solution methods may yet become possible. In this letter, we consider the task of independent component analysis when the independent sources are known to be nonnegative and well grounded, which means that they have a nonzero pdf in the region of zero. It can be shown that in this case, the solution method is basically very simple: an orthogonal rotation of the whitened observation vector into nonnegative outputs will give a positive permutation of the original sources. We propose a cost function whose minimum coincides with nonnegativity and derive the gradient algorithm under the whitening constraint, under which the separating matrix is orthogonal. We further prove that in the Stiefel manifold of orthogonal matrices, the cost function is a Lyapunov function for the matrix gradient flow, implying global convergence. Thus, this algorithm is guaranteed to find the nonnegative well-grounded independent sources. The analysis is complemented by a numerical simulation, which illustrates the algorithm.

Download Full-text

A common neural-network model for unsupervised exploratory data analysis and independent component analysis

IEEE Transactions on Neural Networks ◽

10.1109/72.728398 ◽

1998 ◽

Vol 9 (6) ◽

pp. 1495-1501 ◽

Cited By ~ 15

Author(s):

M. Girolami ◽

A. Cichocki ◽

S.I. Amari

Keyword(s):

Neural Network ◽

Data Analysis ◽

Independent Component Analysis ◽

Network Model ◽

Neural Network Model ◽

Exploratory Data Analysis ◽

Component Analysis ◽

Independent Component ◽

Exploratory Data

Download Full-text

Blind Separation of Excavator Noise Signals in Frequency Domain

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.105-107.723 ◽

2011 ◽

Vol 105-107 ◽

pp. 723-728

Author(s):

Li Da Liao ◽

Qing Hua He ◽

Zhong Lin Hu

Keyword(s):

Independent Component Analysis ◽

Modal Analysis ◽

Frequency Domain ◽

Time Domain ◽

Component Analysis ◽

Independent Component ◽

Blind Separation ◽

Noise Sources ◽

Amplitude Spectra ◽

Complex Valued

In order to identify noise sources of an excavator in non-library environment, a complex-valued algorithm in frequency domain was applied. Firstly, an acoustic camera was used to acquire excavator’s noise signals, which were convolutive mixtures in time domain interfered by echo. Secondly, signals in time domain transformed into frequency domain by FT, turned to be complex-valued mixtures. Then, independent components of noise signals were obtained through separation of complex-valued mixtures using complex-valued algorithm based on independent component analysis. Finally, according to noise of diesel with muffler was mainly consist of surface noise, the relationship between principal frequencies and structrual parts was founded by comparing frequency-amplitude spectra and modal analysis in Ansys. Research shows that complex-valued algorithm based on fast fixed-point independent component analysis can effectively separate noise signals from an excavator in time domain, and noise sources can be well ascertained by comparing the modal analysis with blind separation components.

Download Full-text