Modelling and Verification of Real-Time Systems
Using Timed Automata: Theory and Practice
A dissertation in Computer Systems for the degree of Doctor of
Philosophy. Publicly examined in room X, Uppsala University, 19
February 1999. Technical Report DoCS 99/101. ISSN 0283-0574.
ABSTRACT
During the last decade, model-checking techniques for the verification
of timed systems have been developed based on the theory of timed
automata. The practical limitation in applying these techniques to
industrial-size systems is the huge amount of time and memory needed
to explore and store the state-space of the system model.
In this thesis, we improve the current status of model-checking techniques for timed systems by developing symbolic, on-the-fly and compositional verification techniques for timed automata. A common characteristics of the model-checking techniques presented is that they use efficient constraint-solving techniques to symbolically represent and manipulate the state-space. To avoid construction of the full state-space of the system model two techniques are used: on-the-fly generation of the state-space and a compositional model-checking technique. The memory-usage is further reduced by developing a minimal and canonical data structure for the class of constraints used in the model-checking algorithm, which reduces the size of each individual state. Two other techniques to reduce the total number of states explored and stored during verification are also presented.
The developed techniques have been implemented in the verification
tool UPPAAL. To demonstrate the potential
applications of our model-checking techniques, we present three
industrial-size case studies where the UPPAAL
tool is applied.
THESIS
The whole thesis can be obtained as a gzipped postscript file (.ps.gz) or as a pdf file (.pdf). Parts of the thesis can be
obtained below:
Please also download the errata which contains some minor corrections to the thesis (.ps.gz, .pdf).
Hardcopies of the thesis are available from Department of Computer Systems,
Uppsala University, P.O. Box 325, SE-751 05 Uppsala, Sweden.