M.S. Zakharov - Head, Department of CJSC «Sberbank-Technology»(Moscow)
S.M. Zakharov - Dr.Sc.(Phys.-Math.), Professor, Chief Researcher, OAO Institute of Electronic Control Computers of the I.S. Brooke-s (Moscow)

The efficiency of different types of wavelets under nonstationarity photoplethysmogram spectral analysis of heart rate was studied. Plethysmograms indicated with the thumb of the patient by using a «Biomouse» device represent optoelectronic pair built-in a computer mouse. The recorded signal was digitized with a sampling frequency of ~ 1 kHz and pre- amplified. Duration of the signal registration was approximately one hundred cardiac cycles. Four types of basic maternal functions were used for, two of which being real (Wave, Mhat wavelets), and two complex (Morlet, Rectangular wavelets). Wavelet transformations have basic properties or characteristics of the low frequencies which have a good frequency resolution and the high frequencies which have a poor frequency and good time resolution. The wavelets pairwise equivalent (Wave, Mhat) and (Morlet, Rectangular) is shown. To evaluate the results of spectral analysis it proved to be more preferable to use wavelet of rectangular shape by virtue of a better frequency resolution. This result is directly related to the width of the wavelet used in real and Fourier space. The comparative spectral analysis of photoplethysmograms showed significant differences in the dynamics of the spectrograms of a healthy person and a person with severe pathology.