ISSN 1812-7231 Klin.inform.telemed. Volume 2, Issue 3, 2005, Pages 42-50
V. I. Shulgin, A. V. Morozov, E. V. Volosyuk
"XAI MEDICA", National aerospace university "KhAI", Kharkiv, Ukraine
Blind source separation technology application to biomedical signal processing
The traditional methods of the time, spectral and spaceetime signal processing can be used for the signals extraction from their mixture only in the case when the information about the time and spectral characteristics or the mixing law are available. In the case of absence of such information or its negligibility and similarity of the time and spectral characteristics of mixed signals, the "blind source separation" problem is appear. Its solution is based on statistical indepenn dence of separated signals.
In this work the some problems of biomedical signal processing where the "blind source separation" method can be effectively used, and fundamentals of Bayesian approach to solving such problem are discussed.
Blind Source Separation, BSS, independent component analysis, ICA, Bayesian approach, statistical independence, electroencephalography, electrocardiography, fetal cardiography
Full-text version http://kit-journal.com.ua/en/viewer_en.html?doc/2005_3/9.pdf