Power Efficient and LOw Complexity ApproacheS for Cell FrEe MassIve MIMO Networks – POSEIdON

The next wireless communications networks envisaged for 6G will have to satisfy the ever-increasing demand for connectivity. However, in order to limit the ecological impact of cellular networks, the energy efficiency improvement is mandatory to support the increase in capacity of these systems. Among the solutions considered, distributed antenna systems, and cell free concepts, are serious candidates. These solutions, known as Cell-Free massive MIMO, allow, even for low carrier frequencies, to perform spatial multiplexing, can take advantage of macro diversity and get rid of inter-cell interference, a limiting factor of today's cellular solutions. It is a considerable paradigm shift that can overcome the limitations of existing networks and consider a novel approach adapted to the challenges of tomorrow: performance, flexibility, energy efficiency. Nevertheless, in the present literature, these solutions are still evaluated with theoretical assumptions and models. Therefore, the level of maturity of these techniques remains low.

The POSEIDON project aims at increasing the maturity of these techniques by proposing a fair analysis of the physical and access layers based on signal propagation predictions and the incorporation of hardware impairment models. PHY and MAC algorithms will be developed allowing, in addition to performance optimization, the improvement of energy efficiency in a sector that is becoming more and more energy intensive. Investigated techniques include: grant-free access scheme, AP clustering, index modulation, and scalable distributed processing. They will be optimally designed according to hardware impairments, and possibly powered by AI algorithms. A comparison of distributed and co-localized schemes in terms of performance and energy efficiency will be proposed, integrating hardware imperfections model and connectivity limits (Backhaul). "The comparison will be based on the development of a digital model of the system in two selected environments (possibly: dense urban and factory) for broadband and mMTC use cases. This project, coordinated by CEA-Leti, will draw on the combined expertise of SIRADEL (Engie Group), CentraleSupelec and CNAM. This project proposes to go beyond the existing literature and thus accompanies a still very recent but already very active research dynamic. It opens opportunities for patents, software solutions (testing) and publications in the pre-standardization period of 6G.

The project will allow the partners to (i) improve their skills in topics such as grant free access schemes, channel estimation, low-complexity massive MIMO precoding for cell free telecommunications networks (ii) contribute to the research effort on 6G network and (iii)maintain key skills at the national level. Finally, in addition to a dissemination activity through international journal publications and conferences papers, a particular attention will be paid to the generation of intellectual property. This will also strengthen the partners outreach at national and international levels.