The University of Bremen is a medium-sized German university with around 20,000 students. Bremen offers a wide range of subjects and degrees for its committed and talented students: more than 100 master’s and bachelor’s degrees, as well as the state law exam. Moreover, with research-based learning, the university has reinterpreted project-based courses, a defining feature originating from when the University of Bremen was founded. As part of the European university network YUFE–Young Universities for the Future of Europe it is developing a new model for European higher education together with nine other universities.
The Department of Communications Engineering within the Institute for Telecommunications and High-Frequency Techniques has long-time expertise in conducting research in the areas wireless transmission systems, algorithms, and signal processing techniques. By the participation in DFG programs, relevant EU (Fantastic-5G, METIS, iJOIN) and BMBF (HiFlecs, BZKI, TACNET 4.0, IRLG) funded research projects, and industry projects the gained results in basic research are exploited for the development and standardization of 5G and Beyond 5G mobile communication systems and wireless sensor networks. The department is member in relevant organizations like VDE/ITG specialist group KT1, EU platform NetWorld2020, Arbeitskreis I4.0 Funk and ETSI.
In the FunKI project, the University of Bremen will research very efficient receiver structures from theoretical concept development to prototypical implementation. Starting from the theoretical concept development, the procedures are researched and demonstrated by means of prototypical implementation, taking into account implementation issues and 5G-specific system parameters and scenarios. Two different approaches will be pursued. The Information Bottleneck Method (IBM) is a very general approach for learning information processing methods that is based on information theory. This method has already been used successfully for the optimization of quantizers and for the design of discrete decoders where iterative decoding is implemented very efficiently with the help of learned lookup tables (LUTs). The goal is to extend this approach to more general code structures and to implement the hardware together with the project partners. This allows the analysis of the implementation efficiency and the corresponding message metrics under conditions of realistic hardware implementations. In a second approach, concepts for MIMO equalization and channel decoding are investigated with the help of machine learning techniques. Especially the efficient decoding of short channel codes with high reliability is a challenge and shall be addressed both by learned neural networks and by ML-based adaptation of conventional decoding methods. Based on the identified 5G use cases, an SDR-based 5G transmission link will be built and the AI-based receiver procedures will be implemented. This enables the adaptation of the procedures to real transmission systems and scenarios and proves the practical suitability of AI-based transmission systems.
For more information, pleas visit www.ant.uni-bremen.de
Back to the consortium-page