This paper presents a method of analog fault diagnosis using neural networks. The primary focus of the paper is to provide robust diagnosis using a simple mechanism for automatic test pattern generation while reducing test time. A new diagnosis framework consisting of a white noise generator and an artificial neural network for response analysis and classification is proposed. This approach moves the diagnosis of analog circuits closer to the goal of built-in test. Networks of reasonable dimension are shown to be capable of robust diagnosis of analog circuits including effects due to tolerances.

No abstract

This paper presents experimental results using neural networks to provide gohogo testing and fault diagnosis of analog circuits. The primary focus of the work is to reduce test time and provide a simple mechanism for automatic test pattern generation. Networks of reasonable dimension are shown to be capable of robust diagnosis of analog circuits including effects due to tolerances and nonlinearities. The concepts of the work are extended to include an approach to built-in test of analog or mixed signal ASICs.