RESEARCH ON WAVELET PACKET ALGORITHM FOR FEATURE EXTRACTION OF HEART SOUND SIGNAL

In order to extract pathological features of heart sound signal accurately, an algorithm for extracting the sub-band energy is developed on the basis of the wavelet packet analysis. In the light of the spectrum analysis of heart sound signal, the db6 wavelet, with high energy concentration and good local characteristics, is taken as the mother function , and the optimal wavelet packet basis of heart sound signal is picked out. Then, various heart sound signals are decomposed into four layers and the wavelet packet coefficients of the optimal basis are obtained. According to the equal-value relation in time domain between wavelet packet coefficients and signal energy, the normalized sub-band energy of the optimal basis is extracted as the feature vector. The average value of separability measure is 3.9349 by using the sort separability criterion, which indicates that the algorithm is effective for feature extraction of heart sound signal.