A Survey of Feature Extraction Approaches in Analog Circuit Fault Diagnosis

Feature extraction is the key process in any pattern recognition issues. There is no exception in analog circuit fault diagnosis, because fault diagnosis is equivalent to pattern recognition issue in nature. In this paper, several feature extraction approaches in the field of analog circuits fault diagnosis are summed up. Newly appeared entropy-based, kernel-function-based, fractal-theory-based, rough-set-based feature extractions approaches are described besides the widely used wavelet analysis approach. The advantages and disadvantages of these approaches are discussed also. Potential solutions and developing trends of these feature extraction approaches are indicated.