Diagnosis of Local Spot Defects in Analog Circuits

被引：44

作者：

Huang, Ke ^{[1
]}

Stratigopoulos, Haralampos-G. ^{[2
]}

Mir, Salvador ^{[2
]}

Hora, Camelia ^{[3
]}

Xing, Yizi ^{[3
]}

Kruseman, Bram ^{[3
]}

机构：

[1] Univ Texas Dallas, Richardson, TX 75080 USA

[2] Univ Grenoble 1, TIMA Lab, French Natl Res Council CNRS, Grenoble Inst Technol, F-38031 Grenoble, France

[3] NXP Semicond, NL-5656 Eindhoven, Netherlands

来源：

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT | 2012年 / 61卷 / 10期

关键词：

Analog circuit testing; automobile electronics; failure analysis (FA); fault diagnosis; pattern recognition; FAULT-DIAGNOSIS; SENSITIVITY; CLASSIFIERS;

D O I：

10.1109/TIM.2012.2196390

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

We present a method for diagnosing local spot defects in analog circuits. The method aims to identify a subset of defects that are likely to have occurred and suggests to give them priority in a classical failure analysis. For this purpose, the method relies on a combination of multiclass classifiers that are trained using data from fault simulation. The method is demonstrated on an industrial large-scale case study. The device under consideration is a controller area network transceiver used in automobile systems. This device demands high-quality control due to the reliability requirements of the application wherein it is deployed. The diagnosis problem is discussed by taking into consideration the realities of this case study.

引用

页码：2701 / 2712

页数：12

共 35 条

[1] A modular fault-diagnostic system for analog electronic circuits using neural networks with wavelet transform as a preprocessor [J].