Fault Diagnosis of Analogue Circuits Using Hyper-sphere SVM and AdaBoost

In order to solve the problems of correctly identifying fault classes in fault diagnosis of analogue circuit and improve classification ability, a fault diagnosis method for analog circuits based on AdaBoost algorithm and Hyper-Sphere Support Vector Machine (Hyper-sphere SVM) is developed in this paper. This algorithm uses Hyper-sphere SVMs as weak learners of AdaBoost and use AdaBoost algorithm to improve the accuracy of weak learners. Simulation results of diagnosing an analog circuit show us the proposed technique has the higher classification accuracy compared with Support Vector Machine and has confirmed the validity of the method.