• DocumentCode
    3031350
  • Title

    Incremental Bisimulation Abstraction Refinement

  • Author

    Lei Song ; Lijun Zhang ; Hermanns, Holger ; Godskesen, Jens Chr

  • Author_Institution
    Saarland Univ., Saarbrucken, Germany
  • fYear
    2013
  • fDate
    8-10 July 2013
  • Firstpage
    11
  • Lastpage
    20
  • Abstract
    Abstraction refinement techniques in probabilistic model checking are prominent approaches to the verification of very large or infinite-state probabilistic concurrent systems. At the core of the refinement step lies the implicit or explicit analysis of a counterexample. This paper proposes an abstraction refinement approach for the probabilistic computation tree logic (PCTL), which is based on incrementally computing a sequence of may- and must-quotient automata. These are induced by depth-bounded bisimulation equivalences of increasing depth. The approach is both sound and complete, since the equivalences converge to the genuine PCTL equivalence. Experimental results with a prototype implementation show the effectiveness of the approach.
  • Keywords
    automata theory; bisimulation equivalence; formal verification; probabilistic logic; probability; abstraction refinement approach; abstraction refinement techniques; depth-bounded bisimulation equivalences; explicit analysis; genuine PCTL equivalence; implicit analysis; incremental bisimulation abstraction refinement; infinite-state probabilistic concurrent systems; may-quotient automata; must-quotient automata; probabilistic computation tree logic; probabilistic model checking; refinement step; Abstracts; Algebra; Automata; Computational modeling; Model checking; Probabilistic logic; Safety; Abstraction Refinement; Bisimulation; Concurrent Probabilistic Systems; Probabilistic CTL model checking;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Application of Concurrency to System Design (ACSD), 2013 13th International Conference on
  • Conference_Location
    Barcelona
  • Type

    conf

  • DOI
    10.1109/ACSD.2013.5
  • Filename
    6598336