Repairable Cell-Based Chip Design for Simultaneous Yield Enhancement and Fault Diagnosis

Sheng Lin Lin, Cheng Hung Wu, Kuen Jong Lee

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Fault diagnosis plays a major role in IC yield enhancement. Due to circuit structure and ATPG limitation, there exist many undistinguished fault pairs after applying test patterns and diagnosis patterns, including equivalent fault pairs and aborted fault pairs. This paper proposes a scan-based repair-for-diagnosis architecture that can distinguish undistinguished fault pairs by repairing cell defects. A repairable standard cell design technique is presented that makes the repair of defective cells easy to control. To efficiently distinguish all targeted undistinguished fault pairs, a novel fault-grouping method is developed and applied to the proposed scan-based repair-for-diagnosis architecture. With this architecture, one can distinguish multiple fault pairs and repair those defective cells hence improving yield at the same time. Experimental results show that our proposed architecture can distinguish all targeted undistinguished fault pairs and repair the defective cells with low area overhead.

Original languageEnglish
Title of host publicationProceedings - 2016 IEEE 25th Asian Test Symposium, ATS 2016
PublisherIEEE Computer Society
Pages25-30
Number of pages6
ISBN (Electronic)9781509038084
DOIs
Publication statusPublished - 2016 Dec 22
Event25th IEEE Asian Test Symposium, ATS 2016 - Hiroshima, Japan
Duration: 2016 Nov 212016 Nov 24

Publication series

NameProceedings of the Asian Test Symposium
ISSN (Print)1081-7735

Other

Other25th IEEE Asian Test Symposium, ATS 2016
CountryJapan
CityHiroshima
Period16-11-2116-11-24

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Repairable Cell-Based Chip Design for Simultaneous Yield Enhancement and Fault Diagnosis'. Together they form a unique fingerprint.

Cite this