Identification method of lower limb action pattern based on supporting vector multivariate classification

摘要: An identification method of lower limb action pattern based on supporting vector multivariate classification includes the following steps: myoelectric signal is firstly collected, collected pretreated, time domain and frequency feature extraction carried out, a small mobile window used for sampling when extraction, simultaneously calculation dispersed out in quantum; three block statistical values mean square value, absolute value average variance are selected as space established, obtained by using Mallat decomposition domain, PCA pivot analysis pressure divided into support object set pendular target according to plantar signal, then SVM classifier simplified classifying data sets so that result output. The overcomes disadvantages traditional spectral analysis, provides with good robustness easy mode.