Dipl.-Inform. Michael Kirsten Researcher / PhD Student
Room 228, Building 50.34 Phone: Fax: Email: + 49 721 608 45648 + 49 721 608 44021 kirsten1738421297Xfd4∂kit.edu PGP/GPG: Encrypted and signed emails are strongly encouraged. Key: Fingerprint: 0xFDD82850 76CF 2BCA 6D06 FFD9 4731 36FD 35B9 789A FDD8 2850	KIT  /  Institute of Theoretical Informatics (ITI) Am Fasanengarten 5 Building 50.34 76131 Karlsruhe Germany

Research Interests

• Formalisation and verification of axiomatic/declarative properties for:
  • Analysis of social choice mechanisms (in the broader sense), in particular voting rules
  • (Static) program analysis for information-flow properties
• Software (bounded) model checking
• Deductive program verification

Projects

• Lehre hoch Forschung-PLUS

  

  The main goal of the project Lehre^Forschung-PLUS is to improve the study conditions and to continuously raise the qualitiy of teaching by an extensive development of the curricula for bachelor and master students.

  Our subproject comprises three primary actions, which all aim to implement and strengthen project- and research-oriented, as well as interdisciplinary, teaching in the computer science curricula.

  These comprise the further development and improvement of the project- and research-oriented master module "Research Practice", the project-oriented bachelor module "Software Engineering Practice", as well as introducing the interdisciplinary and research-oriented master profile "Internet and Society".

  Lehre^Forschung-PLUS

• Specification and Verification of Voting Schemes

  

  Voting schemes, as a method to combine individual preferences to an aggregated election result, are part of the fundamental democratic principles. It is thus vital that voting schemes are working as intended. Existing voting schemes have shown undesired and sometimes surprising properties (e.g., negative voting weights in federal elections in Germany) — for more complicated, newly designed voting schemes which become possible due to computer assistance in the vote counting process, these issues are likely to emerge as well.

  The goal of this project is to develop formal verification techniques which allow to check properties of voting schemes without the huge overhead of user interaction needed in full functional verification. The resulting methodology could then be used in an iterative design and implementation process of new voting schemes.

  COST Action IC1205

• DeduSec – Program-level Specification and Deductive Verification of Security Properties

  

  In recent years tremendous progress has been achieved in formal verification of functional properties of computer programs. At the same time seminal papers have been published showing that it is in principle possible to formulate information-flow problems as proof obligations in program logics. The overall goal of the project is to leverage these advances together with our own experience in formal methods for functional properties in order to specify and verify security properties.

  DeduSec Project
  (part of the DFG Priority Programme 1496 Reliably Secure Software Systems – RS³ )

• KeY – Integrated Deductive Software Design

  

  The KeY System is a formal software development tool that aims to integrate design, implementation, formal specification, and formal verification of object-oriented software as seamlessly as possible. At the core of the system is a novel theorem prover for the first-order Dynamic Logic for Java with a user-friendly graphical interface.

  The project was started in November 1998 at the University of Karlsruhe. It is now a joint project of KIT, Chalmers University of Technology, Gothenburg, and TU Darmstadt. The KeY tool is available for download.

  KeY Project

Tools

• CombinedApproach - Automatic and Precise Noninterference Verification of Java Programs
• BEAST - Automated Election Verification via Bounded Model Checking
• KeY - Integrated Deductive Software Design

Teaching

Supervised Projects and Theses

Publications (grouped chronologically)