Analog fault diagnosis using decision fusion

An approach to analog fault diagnosis for circuits with tolerance is presented based on data fusion. The proposed strategy consists of preliminary diagnosis and decision fusion. Preliminary diagnosis is performed separately by using one kind of fault signatures from multiform circuit responses. The preliminary diagnosis results are aggregated using an evidential reasoning algorithm and fault location is accomplished according to the fusion results based on the fusion decision regulation. The experimental results show that the proposed diagnosis approach can produce a more meaningful and accurate diagnosis because of combining the evidences effectively and has the capability to diagnose catastrophic and parametric faults in analog circuits with satisfactory accuracy.