In order to realize dynamic analysis and pattern recognition of knee joint acoustic emission signals, principal component analysis, difference test and classification test based on support vector machine were carried out to study the acoustic emission signals generated by knee joints in different stages of sitting-standing-sitting. The characteristic parameters of the acoustic emission signals were extracted into two principal components after linear changes; the difference test of the acoustic emission signals generated by the knee joint in the two motion stages shows that the result of the progressive significance of the healthy group was the principal component F1< 0. 05, the principal component F2>0. 05, the progressive significance results of the control group was less than 0. 05; The classification accuracy of the support vector machine for the acoustic emission signal of the knee joint reached 97. 9%. The results show that principal component analysis can successfully reduce the dimension of knee acoustic emission signals; the acoustic emission signals at different stages of movement are different, which is particularly obvious in the diseased knee joint; the support vector machine method can accurately diagnose and identify.