The Open Testbed Berlin – 5G and Beyond project, funded by the German Federal Ministry of Education and Research (BMBF) under the program “Selbstbestimmt und sicher in der digitalen Welt” (grant number 16KIS0985), ran from 1 September 2019 to 31 December 2022. It was coordinated by the Technical University of Berlin (TUB) and involved eight partners: Fraunhofer Heinrich‑Hertz‑Institut (HHI), BISDN GmbH, ADVA Optical Networking SE, IAF GmbH, highstreet technologies GmbH, NVIDIA ARC GmbH, and the Berlin Senate Department for Economy, Energy and Employment (SenWEB Berlin). The project was divided into seven work packages (APs); TUB’s Telecommunication Networks (TKN) and Network‑Information‑Theory (NIT) departments led AP 2 (Outdoor RAN) and AP 3 (VLC/Wi‑Fi microcells) and also contributed to AP 1 (open testbed architecture) and AP 6 (application tests and validation).

Technically, the project pursued a flexible cloud‑RAN architecture that integrates hybrid visible‑light‑communication (VLC) and Wi‑Fi microcells into a 5G core network. The microcells were designed for operation at 26 GHz, fabricated as prototypes, and deployed in the Berlin testbed. Their performance was evaluated experimentally, focusing on data‑rate and latency targets required for future mobile applications. The architecture includes a fibre‑based transport network (Ethernet, PON), a software‑defined cloud network, and a distributed core with centralised control. Integration of the VLC/Wi‑Fi nodes into the core was achieved through software‑defined networking (SDN) interfaces, enabling seamless handover and resource management.

Parallel to the microcell work, TUB NIT developed GPU‑accelerated solutions for millimetre‑wave access networks. The research concentrated on 5G non‑orthogonal multiple access (NOMA) and full‑duplex techniques, both of which demand massive parallelism and low‑latency processing at the physical layer. By leveraging modern graphics processors and optimised GPU software libraries, the team aimed to reduce decoding complexity and improve real‑time performance. Experimental results demonstrated that GPU acceleration can meet the stringent latency requirements of 5G‑NOMA and full‑duplex operations, although specific throughput figures were not disclosed in the report.

The project’s outcomes include a validated prototype of a hybrid VLC/Wi‑Fi microcell, a demonstrator of GPU‑accelerated PHY‑layer processing for mmWave links, and a comprehensive testbed architecture that supports future 5G and beyond deployments. The findings are expected to enhance the economic viability of high‑capacity, low‑latency wireless networks and to provide a scientific basis for further research in integrated optical‑wireless systems. Dissemination of the results occurred through conference papers, journal articles, and graduate theses, underscoring the project’s contribution to both academia and industry.