Teaching Activities - Summer 2024
Notice:
We announce any news with respect to teaching through stud.ip. Enroll for the respective courses on stud.ip for information!
- Repetitorium: Informatik für Ingenieur*innen - Einführung und Überblick (Bachelor)
Eintragung in stud.ip.
Das Repetitorium ist für Teilnehmer*innen der Vorlesung aus dem Wintersemester. Der Fokus liegt hier auf eigenständiger Programmierung. Die Folien zur Theorie stehen weiter als Folien aus dem Sommersemester zur Verüfgung. Fragen dazu können jederzeit in der wöchentlichen Plenumsveranstaltung behandelt werden. - Lecture (Master IIW, CS, ICS): Modelchecking - Proof Engines and Algorithms
Subscribe in stud.ip.
Correctness is a major concern in embedded systems. Model checking can fully automatically prove formal properties about digital hardware or software. Such properties are given in temporal logic, e.g., to prove "No two orthogonal traffic lights will ever be green at the same time."
But what are the limitations of model checking?
How are the models generated from a given design?
And how do the underlying reasoning algorithms work so effectively in practice despite a computational complexity of NP hardness and beyond?
The lecture will answer these questions. Open source tools will be used to gather a practical experience.
Further information is available in stud.ip (use link above). - Lecture (Master IIW, CS, ICS, MM, MEC): Designing Dependable Systems
Subscribe in stud.ip.
The term dependability summarizes the concepts reliability, availability, maintainability, safety, and security. All of these aspects are interdependent and are not directly encompassed by the functional specification for a system. This makes the design of dependable system difficult even though all embedded or cyber physical systems have to comply with certain dependability requirements.
This lecture covers how to analyze and quantize dependability with a focus on reliability and availability. Moreover, design techniques for dependable systems are covered. This includes general concepts like redundancy as well as techniques chosen for specific levels in the system hierarchy, e.g., digital hardware design or dependable software.