PCA-based parametric fault analysis for analog integrated circuits

The wavelet filtering technology and principal components analysis (PCA) is used to extract the signatures of circuits in this paper. Firstly, the output signal of the analog circuits is filtered by Harr wavelet packet. Then, the energetic values in eight subbands are extracted as the principal components. By comparison with the eigenvalues of the energetic relative matrices, different states of the circuits are identified. The idea is discussed and the result is given in the paper.