Short-term analysis of heart-rate variability by adapted wavelet transforms

Wavelet packets (WP) and cosine packets (CP) algorithms are utilized to characterize autonomic nervous system (ANS) modulation of cardiovascular activity in order to better understand short-term cardiovascular regulation. The packets have the ability to localize changes in the characteristics of nonstationary heart rate variability without loosing time and frequency information. Since the WP and CP methods are free of any assumptions about the stationarity of the biological signals, they are preferred in signal processing over other parametric modeling techniques. Furthermore, the WP and CP methods do not require the estimation or assumption of the number of model parameters.