Bisimilarity on basic process algebra is in 2-ExpTime (an explicit proof)
From MaRDI portal
Abstract: Burkart, Caucal, Steffen (1995) showed a procedure deciding bisimulation equivalence of processes in Basic Process Algebra (BPA), i.e. of sequential processes generated by context-free grammars. They improved the previous decidability result of Christensen, H"uttel, Stirling (1992), since their procedure has obviously an elementary time complexity and the authors claim that a close analysis would reveal a double exponential upper bound. Here a self-contained direct proof of the membership in 2-ExpTime is given. This is done via a Prover-Refuter game which shows that there is an alternating Turing machine deciding the problem in exponential space. The proof uses similar ingredients (size-measures, decompositions, bases) as the previous proofs, but one new simplifying factor is an explicit addition of infinite regular strings to the state space. An auxiliary claim also shows an explicit exponential upper bound on the equivalence level of nonbisimilar normed BPA processes. The importance of clarifying the 2-ExpTime upper bound for BPA bisimilarity has recently increased due to the shift of the known lower bound from PSpace (Srba, 2002) to ExpTime (Kiefer, 2012).
Recommendations
Cited in
(11)- Strong bisimilarity of simple process algebras: Complexity lower bounds
- Bisimilarity on basic parallel processes
- Equivalence of pushdown automata via first-order grammars
- Nested session types
- Bisimulation equivalence and regularity for real-time one-counter automata
- Game characterization of probabilistic bisimilarity, and applications to pushdown automata
- Branching bisimilarity of normed BPA processes is in NExpTime
- Branching bisimilarity on normed BPA is EXPTIME-complete
- Two lower bounds for BPA
- Deciding the bisimilarity of context-free session types
- System \(F^\mu_\omega\) with context-free session types
This page was built for publication: Bisimilarity on basic process algebra is in 2-ExpTime (an explicit proof)
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4913773)