Fault diagnosis in analog circuits using multiclass Relevance Vector Machine

In this paper the capability of Relevance Vector Machines to perform multiclass classification has been illustrated. It has been demonstrated how faults in an analog circuit can be diagnosed by analyzing these faults as a multiclass machine learning problem. A simple first order Op-amp RC circuit va...

Full description

Saved in:
Bibliographic Details
Published in2011 International Conference on Emerging Trends in Electrical and Computer Technology pp. 641 - 643
Main Authors Jain, V, Pillai, G N, Gupta, I
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.03.2011
Subjects
Analog circuits
Bayesian methods
Circuit faults
Machine learning
Mathematical model
Support vector machine classification
Testing
Online AccessGet full text

Cover

More Information
Summary:In this paper the capability of Relevance Vector Machines to perform multiclass classification has been illustrated. It has been demonstrated how faults in an analog circuit can be diagnosed by analyzing these faults as a multiclass machine learning problem. A simple first order Op-amp RC circuit validates our methodology which can be further applied to more complex analog circuits employing a larger number of electronic components.
ISBN:1424479231
9781424479238
DOI:10.1109/ICETECT.2011.5760196