Iterative Abstraction using SAT-based BMC with Proof Analysis

  • Authors:
  • Aarti Gupta;Malay Ganai;Zijiang Yang;Pranav Ashar

  • Affiliations:
  • NEC Laboratories America, Princeton, NJ;NEC Laboratories America, Princeton, NJ;NEC Laboratories America, Princeton, NJ;NEC Laboratories America, Princeton, NJ

  • Venue:
  • Proceedings of the 2003 IEEE/ACM international conference on Computer-aided design
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

Resolution-based proof analysis techniques have been proposedrecently to identify a sufficient set of reasons for unsatisfiabilityderived by a CNF-based SAT solver. We have adapted thesetechniques to work with a hybrid SAT solver. We use the proofanalysis technique with SAT-based BMC, in order to generateuseful abstract models. Our abstraction procedure is usediteratively in a top-down framework, starting from the concretedesign, where we apply BMC on increasingly more abstractmodels. We apply various SAT-based and BDD-basedverification methods on these abstract models, in order to obtainproofs of correctness, or to perform deeper searches forcounterexamples. We demonstrate the effectiveness of ourprototype implementation on several large industry designs.