Fault detection and classification in linear integrated circuits: an application of discrimination analysis and hypothesis testing

Epstein, Benjamin R.; Czigler, M.; Miller, Steven

doi:10.1109/43.184847

Cited by 66 publications

(33 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…It has been suggested, therefore, that simple thresholds are insufficient for fault detection and statistical methods should be employed [3]. Linear discriminant analysis (LDA), which is a classical statistical method [4], was, therefore, applied in 1993 [3].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Applying a robust heteroscedastic probabilistic neural network to analog fault detection and classification

Yang

Zwoliński

Chalk

et al. 2000

IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst.

View full text Add to dashboard Cite

Abstract-The problem of distinguishing and classifying the responses of analog integrated circuits containing catastrophic faults has aroused recent interest. The problem is made more difficult when parametric variations are taken into account. Hence, statistical methods and techniques such as neural networks have been employed to automate classification. The major drawback to such techniques has been the implicit assumption that the variances of the responses of faulty circuits have been the same as each other and the same as that of the fault-free circuit. This assumption can be shown to be false. Neural networks, moreover, have proved to be slow. This paper describes a new neural network structure that clusters responses assuming different means and variances. Sophisticated statistical techniques are employed to handle situations where the variance tends to zero, such as happens with a fault that causes a response to be stuck at a supply rail. Two example circuits are used to show that this technique is significantly more accurate than other classification methods. A set of responses can be classified in the order of 1 s.

show abstract

Section: Introductionmentioning

confidence: 99%

“…Linear discriminant analysis (LDA), which is a classical statistical method [4], was, therefore, applied in 1993 [3]. Because of the limitations of LDA, Epstein suggested that neural networks should be applied to the area of fault detection and classification [3].…”

Section: Introductionmentioning

confidence: 99%

Applying a robust heteroscedastic probabilistic neural network to analog fault detection and classification

Yang

Zwoliński

Chalk

et al. 2000

IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst.

View full text Add to dashboard Cite

show abstract

“…Please note that this extension of the definition of a faulty circuit is based on similar considerations as in [8,9,23]. In the approach proposed in this paper, these rather crude test criteria take effect only when no given specification induces a reasonable limit for permissible parameter deviations.…”

Section: Bounding the Acceptance Regionmentioning

confidence: 99%

“…In order to approximate the acceptance region A m , the following proceeding is performed. Based on discrimination analysis a test criterion of the form [7,9,14] t (m) = 0 + T m 0 (1) is computed for each given specification. Thus, linear approximations of the separating hyperplanes are determined.…”

Section: Starting Point and Basic Relationshipsmentioning

confidence: 99%

“…In this situation, statistical methods for decision making are often applied to derive test criteria for the detection of parametric faults based on preferred measurements [7,8,9]. In order to obtain an applicable test design, the performance of the determined test criteria has to be investigated, not only for the nominal case but also in the presence of measurement noise.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Design of robust test criteria in analog testing

Lindermeir¹

Proceedings of International Conference on Computer Aided Design

View full text Add to dashboard Cite

show abstract

Estimation of circuit output measurements including statistically dependent parameters

Papakostas

Hatzopoulos

2003

Circuit Theory & Apps

View full text Add to dashboard Cite

SUMMARYAn extension to the method for estimating statistical variables of output measurements, in order to take into account the statistical variations and matching dependencies between circuit parameters, is studied. The enhanced method results in su cient estimations, while additional circuit simulations are needed. Comparative results from the application of the method and of the Monte-Carlo analysis on three analogue circuits are presented to show the e ectiveness of the extended method.

show abstract

Fault detection and classification in linear integrated circuits: an application of discrimination analysis and hypothesis testing

Cited by 66 publications

References 5 publications

Applying a robust heteroscedastic probabilistic neural network to analog fault detection and classification

Applying a robust heteroscedastic probabilistic neural network to analog fault detection and classification

Design of robust test criteria in analog testing

Estimation of circuit output measurements including statistically dependent parameters

Contact Info

Product

Resources

About