Skip to main navigation Skip to search Skip to main content

An automated fault injection technique based on VHDL syntax analysis and stratified sampling

  • Weiguang Sheng*
  • , Liyi Xiao
  • , Zhigang Mao
  • *Corresponding author for this work
  • Harbin Institute of Technology

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Soft errors, a concern for space applications in the past, became a critical issue in deep sub-micron VLSI design for the continuous technology scaling. An automated fault injection technique is employed to characterise the recovery coverage and soft error sensitivity of the VHDL based design. Lexical and syntax analysis technique is developed to perform automated fault injection task. Stratified sampling technique is used to reduce the fault injection overheads. A fault injector, HSECT (HIT Soft Error Characterization Toolkit) is developed and 3,000 soft errors are injected into a simple RISC processor, DP32-processor. The recovery coverage and soft error sensitivity of the processor are also further investigated to direct the future design of the faulttolerant and dependable circuit.

Original languageEnglish
Title of host publicationProceedings - 4th IEEE International Symposium on Electronic Design, Test and Applications, DELTA 2008
Pages587-591
Number of pages5
DOIs
StatePublished - 2008
Event4th IEEE International Symposium on Electronic Design, Test and Applications, DELTA 2008 - Hong Kong, SAR, Hong Kong
Duration: 23 Jan 200825 Jan 2008

Publication series

NameProceedings - 4th IEEE International Symposium on Electronic Design, Test and Applications, DELTA 2008

Conference

Conference4th IEEE International Symposium on Electronic Design, Test and Applications, DELTA 2008
Country/TerritoryHong Kong
CityHong Kong, SAR
Period23/01/0825/01/08

Fingerprint

Dive into the research topics of 'An automated fault injection technique based on VHDL syntax analysis and stratified sampling'. Together they form a unique fingerprint.

Cite this