🤖 AI Summary
This work proposes the first model checking approach for CTL* temporal logic tailored to infinite families of finite-state transition systems generated by highly configurable systems or software product lines. The method employs context-free graph grammars to uniformly model the entire system family and introduces a grammar-rule-based, compositional state labeling algorithm. This algorithm recursively propagates state labels using only finite contextual information, enabling the verification of whether all, some, or infinitely many members of the family satisfy a given CTL* property. Experimental results demonstrate that the approach effectively supports unified formal verification across infinite system families, offering a scalable solution for reasoning about complex configurable systems within a rigorous logical framework.
📝 Abstract
We study model checking algorithms for infinite families of finite-state labeled transition systems against temporal properties written in CTL*. Such families arise, for example, as models of highly configurable systems or software product lines. We model families using context-free graph grammars. We then develop a state labeling algorithm that works compositionally on the grammar's production rules with limited information about the context in which the rule is applied. The result is a graph grammar modeling the same family but with extended labels. We leverage this grammar to decide whether all, some, or (in)finitely many members of a family satisfy a given temporal property. We have implemented our algorithms and present early experiments.