In 2008, the ACM awarded the prestigious Turing Award - the Nobel Prize in Computer Science - to the pioneers of Model Checking: Ed Clarke, Allen Emerson, and Joseph Sifakis.

 

Why? Because model checking has evolved in the last twenty-five years into a widely used verification and debugging technique for both software and hardware.

 

It is used (and further developed) by companies and institutes such as IBM, Intel, NASA, Cadence, Microsoft, and Siemens, to mention a few, and has culminated in a series of mostly freely downloadable software tools that allow the automated verification of, for instance, C#-programs or combinational hardware circuits.

Subtle errors, for instance due to multi-threading, that remain undiscovered using simulation or peer reviewing can potentially be revealed using model checking.  Model checking is thus an effective technique to expose potential design errors and improve software and hardware reliability.

 

But how does it work, that is, what are its underlying principles?

That's exactly the focus of this lecture series!

 

We will show that model checking is based on well-known paradigms from automata theory, graph algorithms, logic, and data structures.  Its complexity is analyzed using standard techniques from complexity theory.

Model checking is based on checking models. So, we first start by explaining what models are, and will make clear that so-called labeled transition systems, a model that is akin to automata, are suitable for modeling sequential, as well as multi-threading programs. 

This is followed by a detailed study on various classes of system properties such as safety, liveness, and fairness.  It will be shown how finite automata as well as variants thereof that accept infinite words are suitable for verifying regular properties.

The linear-time and branching time temporal logics LTL and CTL are then introduced and compared. Model checking algorithms for these logics together with detailed complexity considerations are covered. 

Finally, abstraction techniques based on bisimulation and simulation relations are covered. These techniques are at key techniques to combat the state explosion problem.

This course is suitable for Hauptdiplom, bachelor (this course is part of the "Wahlkatalog" in theoretical computer science), as well as master students. 

We expect students to have some basic knowledge in:

• Automata Theory
• Mathematical Logic
• Discrete Mathematics
• Complexity Theory
• Algorithms and Data Structures

The course book (see below) will contain a summary of the issues in these areas that are relevant to this lecture series.  We believe that this course is also well-suited for students in electrical engineeering and mathematics.

This course provides the basis for the follow-up courses:

Advanced Model Checking (planned for SS 2009)
Model Checking Labs (planned for SS 2009)

Moreover, it is related to courses such as:

• Semantics and Verification of Software
• Automata and Reactive Systems (i7)
• Automata on Infinite Words (i7)
• Formal Methods for Embedded Systems (i11)

There are thus plenty of opportunities to combine this course with others!  We also regularly offer seminars that are based on model checking and offer various master and diplom theses on model checking and its applications.

The course will basically cover Chapters 1 through 7 (upto section 7.3).

Additional literature:

• E.M. Clarke, O. Grumberg, D.A. Peled: Model Checking, MIT Press, 1999

• M. Huth and M.D. Ryan: Logic in Computer Science - Modelling and Reasoning about Systems, Cambridge University Press, 2nd edition, 2004