Feature Extraction and Recognition of Analog Circuit Fault Signal Based on Wavelet Transform and Partial Singular Values Decomposition

DOI

10.2991/citcs.2012.142How to use a DOI?

Keywords

wavelet transform; singular value decomposition; feature extraction; analog circuit

Abstract

A new algorithm for the effective characteristics' extraction of analog circuit fault signal is proposed with consideration of its complexity. The proposed method is based on wavelet transform and partial singular values decomposition. Firstly, the analog circuit fault signal is decomposed into different scales and construct the initial feature matrix with wavelet coefficients, then divide the matrix into different sub-matrix, calculate the largest singular value of each sub-matrix, and the eigenvectors are constructed by the largest singular value. Then the support vector machine (SVM) with the constructed eigenvectors as its input eigenvectors is employed to recognize the analog circuit signal of different fault mode. The results show that the characteristics of analog circuit signal are efficiently extracted and the recognition precision of SVM is also improved, the new algorithm is a feasible way to process analog circuit fault signal.

Copyright

© 2012, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Si-yang Liang
AU  - Jian-hong Lv
PY  - 2012/11
DA  - 2012/11
TI  - Feature Extraction and Recognition of Analog Circuit Fault Signal Based on Wavelet Transform and Partial Singular Values Decomposition
BT  - Proceedings of the 2012 National Conference on Information Technology and Computer Science
PB  - Atlantis Press
SP  - 553
EP  - 556
SN  - 1951-6851
UR  - https://doi.org/10.2991/citcs.2012.142
DO  - 10.2991/citcs.2012.142
ID  - Liang2012/11
ER  -