Application of the support vector machine to the identification of human pulse signals

Using the Mallat fast algorithm with sym5 wavelet, the pulse waves of 20 heroin druggers and 20 healthy normal subjects are decomposed into two levels. The squared distances from the third and tenth scale coefficients in the second-level decomposition of every pulse wave to the global mean value are used to form a feature vector. The extracted feature vectors have good separable characteristics in a two-dimensional plane. These 40 feature vectors are then used as training samples for designing the network of support vector machine. The network can successfully recognize 38 feature vectors. Using other 10 feature vectors of healthy normal subjects to test the generalization ability of the designed network, all of these vectors are correctly identified. The research result shows that the designed network of the support vector machine has good classification characteristics, generalization ability and some values in the identification of the pulse signals for heroin druggers.