๐ค AI Summary
This work addresses the limited support for O-RAN standards in existing 5G testbeds, which often fail to balance flexibility and open-source accessibility. The authors present an end-to-end, modular O-RAN testbed that, for the first time, integrates the open-source SRS O-CU/O-DU with the commercially compliant Massive Beams MODRAD-SC O-RU. The platform supports flexible switching between Split 7.2a and Split 8 architectures and is compatible with hybrid digital-analog beamforming. Built upon a software-defined radio (SDR) framework, it fully implements a 3GPP- and O-RAN-compliant 5G gNodeB protocol stack, offering high programmability and scalability. This testbed effectively enables validation of 5G algorithms, system demonstrations, and cutting-edge experimentation toward 6G research.
๐ Abstract
In this paper, we present a modular open radio access network (O-RAN) consisting of the 5G Core, a central (O-CU) and distributed unit (O-DU) by Software Radio Systems (SRS) and an O-RAN radio unit (O-RU), MODRAD-SC, by Massive Beams (MB). OCUDU provides an open source 5G-compliant O-CU and O-DU solution developed by SRS, while MB's radio unit is a fully O-RAN compliant category A O-RU. According to O-RAN split 7.2a, OCUDU performs higher layer functions up to the high physical (PHY) layer, while the O-RU handles low PHY and RF functions. This results in an O-RAN-compliant 5G gNodeB. In an alternative configuration, OCUDU and MODRAD-SC operate in a software-defined radio fashion corresponding to split 8, facilitating non-real-time experiments among others. In both cases, the system provides full control over O-CU, O-DU, and O-RU. In addition, we will discuss the possibility to attach an analog beamformer to the O-RU, enabling hybrid digital-analog beamforming. The flexibility and modularity offered by OCUDU and MODRAD-SC enable the practical realization of a multitude of applications, ranging from 5G demonstrators to pre-6G experiments. The system addresses the requirements of academia and industry and is well-suited as an easy-to-use platform for experimental and practical deployments.