No abstract

No abstract

Real-time systems must be properly validated and verified before their manufacturing and deployment in order to increase their reliability and reduce their maintenance cost. Models have been used for a long time to build complex systems, in virtually every engineering field. This is because they provide invaluable help in making important design decisions before the system is implemented. In this paper, the authors propose an approach based on model transformation to apply formal verification techniques to demonstrate the correctness of system designs. At the first step, they describe real-time systems by state chart (machine) diagrams, as source models to generate RT-Maude models (target models). The second step is to use the result models to verify the real-time systems against specified LTL properties using Maude LTL Model-Checker. This approach is illustrated through an example.

No abstract

The work presented in this paper lies in the context of implementing supporting tools for a domain-specific language named SosADL, targeted at the description and analysis of architecture for systems of systems. While the language has formal definition rooted in the Cc-pi calculus, we have adopted the Eclipse ecosystem, including EMF, Ecore and Xtext for the convenience they provide in implementation tasks. Proof-carrying code is a well-known approach to ensure such an implementation involving non-formal technologies conforms to its formal definition, by making the implementation generate proof in addition to usual output artifacts. In this paper, we therefore investigate for an infrastructure that eases the development of proof-carrying code for an Eclipse/EMF/Ecore/Xtext-based tool in relation with the Coq proof assistant. At the core of our approach, we combine an automatic transformation of a metamodel into a set of inductive types, in conjunction with a second transformation of model elements into terms. The first one, reused from our previous work, provides necessary abstract syntax definitions such that the formal definition of the language can be mechanized using Coq. The second transformation is part of the proof generator.