P-complete

Strong Bisimilarity in Transition Systems (SBTS)

Input

An encoding of a finite labeled transition system $N$, and two designated states $p$ and $q$.

Problem

Are $p$ and $q$ strongly bisimilar?

Definitions

A finite labeled transition system is a triple $M = \langle Q, \Sigma, T \rangle$, where $Q$ is a finite set of states, $\Sigma$ is a finite alphabet, and $T \subseteq Q \times \Sigma \times Q$ is the set of transitions. A relation $S \subseteq Q \times Q$ is a strong bisimulation of $M$ if $(s_1, s_2) \in S$ implies for all $x \in \Sigma$ that

  1. whenever $(s_1, x, t_1) \in T$, then for some state $t_2$, $(s_2, x, t_2) \in T$ and $(t_1, t_2) \in S$; and
  2. whenever $(s_2, x, t_2) \in T$, then for some state $t_1$, $(s_1, x, t_1) \in T$ and $(t_1, t_2) \in S$.

The strong bisimulation relation is defined as the union of all strong bisimulations of $M$. States $p$ and $q$ are strongly bisimilar if $(p, q)$ is in the strong bisimulation relation of $N$.

Status

$\P$-complete (Àlvarez et al [1], Balcázar, Gabarró, and Sántha [2]).

Remarks

The reduction to show hardness is from SAM2CVP, and in it the transition system is nondeterministic. It is an open question whether the problem remains $\P$-complete in the deterministic case [2], see SBDTS. Observation Equivalence and Observation Congruence are two related problems that are both shown $\P$-complete in [2]. Zhang and Smolka implement parallel algorithms for equivalence checking [3].

References

  1. C. `Alvarez, J. L. Balcázar, J. Gabarró, and M. Sántha (1991). Parallel complexity in the design and analysis of concurrent systems.
  2. J. L. Balcázar, J. Gabarró, and M. Sántha (1992). Deciding bisimilarity is P-complete.
  3. S. Zhang and S. A. Smolka (1990). Towards efficient parallelization of equivalence checking algorithms.
  4. T. Shoudai (1992). A P-complete language describable with iterated shuffle.
  5. Greenlaw, R., Hoover, H. J., and Ruzzo, W. L. (1995). Limits to Parallel Computation: P-Completeness Theory. Oxford University Press. Problem A.7.10.