Teilprojekt: Relevance based cell-free massive MIMO and in-X interference estimation (ReMI)

Initial Situation

Innovative communications systems are key technologies for digitization and pave the way for a hyperconnected economy and society. The future mobile communications standard 6G will play an important role in this. Before market launch, which is expected around 2030, numerous technological and social issues still need to be resolved. It is already clear that simplified interaction between people and technology will be central to the development of 6G. Crucial to a functioning, hyperconnected world in which "everything interacts with everything" are powerful, trustworthy and sustainable 6G systems that take into account the European Union's principles of action and values from the ground up. In international competition, it is therefore important for Germany to drive forward 6G research early and quickly in order to be able to exert a significant influence on the standardization of 6G.

Objective

6G-ANNA aims to provide a holistic system approach for 6G mobile radio systems. The aim is to develop an overarching understanding of how future requirements on the user side can be addressed with technological innovations. This deep understanding of the requirements, the exploration of the technological concepts as well as the pre-competitive consensus building form the basis for German and European companies to take a leading role in the later standardization and market introduction.

Research Contribution of the University of Bremen

In 6G-ANNA, the University of Bremen addresses two: distributed interference estimation in 6G wide area networks with in-X networks and relevance-based signal processing for cell-free massive MIMO. In centralized interference estimation, the challenge is that 6G networks may consist of complex subnetworks that are not fully centrally controlled. The detection of the interference situation over a mobile cell has to be distributed by terminals and reported to the base station. The focus of this sub-project is therefore AI or machine learning methods for the design, learning and application of a distributed interference estimation procedure. For uplink signal processing of cell-free massive MIMO, the challenge is to transmit and process the raw data multiple radio units that dynamically form the cell for a user and are processed centrally. Therefore, the focus is on relevance-based signal processing algorithms to achieve optimal end-to-end transmission for variable and different fronthauls (e.g. wireless fronhaul, moving RU). Hybrid methods are used that combine classical message engineering methods with machine learning methods to achieve flexible trade-offs between complexity and performance.

Further Information

Details