Title :
Neural network approach to fault diagnosis in CMOS opamps with gate oxide short faults
Author :
Yu, Son-Cheol ; Jervis, B.W. ; Bell, I.M. ; Hall, A.G. ; Taylor, G.
Author_Institution :
Sch. of Eng. Inf. Technol., Sheffield Hallam Univ.
fDate :
4/29/1994 12:00:00 AM
Abstract :
Faults owing to gate oxide shorts in a CMOS opamp have been diagnosed in simulations using artificial neural networks to identify corresponding variations in supply current. Ramp and sinusoidal signals gave fault diagnostic accuracy of 67 and 83%, respectively. Using both test signals 100% diagnostic accuracy was achieved
Keywords :
CMOS integrated circuits; feedforward neural nets; integrated circuit testing; linear integrated circuits; operational amplifiers; pattern recognition; CMOS operational amplifiers; artificial neural networks; fault diagnosis; fault diagnostic accuracy; gate oxide short faults; pattern recognition; ramp signals; simulations; sinusoidal signals; supply current variations; three-layer multilayer perceptron;
Journal_Title :
Electronics Letters
DOI :
10.1049/el:19940472
