RAIRO-Theor. Inf. Appl.
Volume 47, Number 1, January-March 20136th Workshop on Fixed Points in Computer Science (FICS'09)
|Page(s)||69 - 96|
|Published online||10 January 2013|
On characteristic formulae for Event-Recording Automata∗
Laboratoire Bordelais de Recherche en Informatique (LaBRI),
Université Bordeaux 1, 351 cours de
la Libération, 33405
2 Laboratoire d’Informatique Fondamentale (LIF), Aix-Marseille Université, 39 avenue Joliot-Curie, 13453 Marseille Cedex 13, France.
Accepted: 28 September 2012
A standard bridge between automata theory and logic is provided by the notion of characteristic formula. This paper investigates this problem for the class of event-recording automata (ERA), a subclass of timed automata in which clocks are associated with actions and that enjoys very good closure properties. We first study the problem of expressing characteristic formulae for ERA in Event-Recording Logic (ERL ), a logic introduced by Sorea to express event-based timed specifications. We prove that the construction proposed by Sorea for ERA without invariants is incorrect. More generally, we prove that timed bisimilarity cannot in general be expressed in ERL for the class of ERA , and study under which conditions on ERA it can be. Then, we introduce the logic WTμ , a new logic for event-based timed specifications closer to the timed logic ℒν that was introduced by Laroussinie, Larsen and Weise. We prove that it is strictly more expressive than ERL , and that its model-checking problem against ERA is EXPTIME -complete. Finally, we provide characteristic formulae constructions in WTμ for characterizing the general class of ERA up to timed (bi)similarity and study the complexity issues.
Mathematics Subject Classification: 03B44 / 68Q60
Key words: Timed Logic / bisimulation / event-clock automata
© EDP Sciences 2013
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.