Independent Component Analysis of Parameterized Electrocardiogram Signals

Independent component analysis (ICA) of measured signals yields the independent sources, given certain fulfilled requirements. Properly parameterized signals may provide a better view to the considered system aspects, while reducing the amount of data. It is little acknowledged that appropriately parameterized signals may be subjected to ICA, yielding independent components displaying more clearly the investigated properties of the sources. In this paper, we propose ICA of parameterized signals, and demonstrate the concept with a set of electrocardiogram (ECG) measurements and their selected parameterizations