In this module students become familiar with current advanced topics in theoretical communications engineering. This includes current methods and tools from statistical signal processing and statistical and information theoretical modelling of communication systems (e.g. arbitrarily varying channels, copula) and the analysis and design of communication systems using learning algorithms (reinforcement learning, deep neural networks, etc.). The module enables students to deal with current research questions in theoretical communications engineering using modern solid methods.
- Abstract stochastic modeling of communication channels
- Performance analysis of communication systems
- Coding and transmission via arbitrarily variable channels
- Multi-party networks and statistically dependent channels
- Bayesian Inference and Bayesian Statistics
- Fisher Information and Cramer Rao Bound
- Deep Neural Networks and global optimization
- Reinforcement Learning for optimization of complex communication systems
There is an exercise for the lecture. The exercise is conducted as a so-called "Reading Class", in which students present current publications on the above-mentioned topics in the form of short presentations.
Lecturer: Prof. Eduard Jorswieck
Assistant: Pin-Hsun Lin Ph.D., Dr. Bile Peng