The fault inspection and diagnosis of vehicle transmission plays a important role of safety traffic and knocking the traffic accidents down.For the fault character of vehicle continuously variable transmission（CVT）

the fault signal character extract entropy method based on the theory of mutual information entropy and the multiclass classification support vector machine（SVM） algorithm to be easily implemented to deal with the problem of classification for fault state are proposed.Experimental results show that the combination mutual information entropy with multiclass classification SVM algorithm in fault diagnosis of vehicle CVT is the feasibility and efficiency

and the reliability of fault inspection is improved.