A novel method for the fault detection of analog integrated circuits

A novel method for the fault detection of analog circuits is proposed. Simple measurements are used to detect the possible faults in an analog circuit. Bayes decision rule is applied to combine the priori information and the information from testing. A sequential method to evaluate the value of the priori probability is given. Principle component analysis is applied for the calculation of the discrimination function in the case of the measurements being dependent. Examples are given to demonstrate the efficiency and the effectiveness of the algorithm.<
>