Please use this identifier to cite or link to this item:
|Title:||Symbolic model-checking of stateful timed CSP using BDD and digitization|
|Source:||Nguyen, T.K.,Sun, J.,Liu, Y.,Dong, J.S. (2012). Symbolic model-checking of stateful timed CSP using BDD and digitization. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 7635 LNCS : 398-413. ScholarBank@NUS Repository. https://doi.org/10.1007/978-3-642-34281-3_28|
|Abstract:||Stateful Timed CSP has been recently proposed to model (and verify) hierarchical real-time systems. It is an expressive modeling language which combines data structure/operations, complicated control flows (modeled using compositional process operators adopted from Timed CSP), and real-time requirements like deadline and within. It has been shown that Stateful Timed CSP is equivalent to closed timed automata with silent transitions, which implies that the timing constraints of Stateful Timed CSP can be captured using explicit tick events, through digitization. In order to tackle the state space explosion problem, we develop a BDD-based symbolic model checking approach to verify Stateful Timed CSP models. Due to the rich language features, BDD-based system encoding and verification is highly nontrivial. In this work, we show how to systematically encode Stateful Timed CSP models in BDD. Our approach consists of two steps. The first step is to identify maximum primitive components of a given system and then generate finite state machines (FSMs) from them, applying a set of symbolic firing rules. These FSMs are then encoded in the standard way. The second step is to compose the encoded components using a set of BDD-based compositional functions. The proposed method has been implemented in the PAT model checker. It supports properties like reachability, linear temporal logic, etc. The effectiveness of our technique is evaluated with benchmark systems. © 2012 Springer-Verlag.|
|Source Title:||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|Appears in Collections:||Staff Publications|
Show full item record
Files in This Item:
There are no files associated with this item.
checked on Dec 13, 2017
checked on Dec 9, 2017
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.