Two presentations showcasing recent results on in-depth research conducted on RAN energy savings will be presented at the upcoming O-RAN Alliance Face-to-Face Meeting taking place in Dallas, Texas on October 27-31, 2025.
Supported by funding from the NTIA, the O-RAN Energy Efficiency Research Project (under the Public Wireless Supply Chain Innovation Fund) is an initiative aimed at advancing sustainable, open, and data-driven approaches to next-generation wireless networks.
The presentations to be featured at the O-RAN Alliance Face-to-Face Meeting are:
“Energy Efficiency Testing and Power Modeling of O-RAN Radio Units”
This talk introduces a versatile testing and modeling framework developed under the POET (Platform for O-RAN Energy Efficiency Testing) initiative. This work delivers one of the most comprehensive open datasets to date on multi-vendor O-RU power measurements, coupled with a validated component-level energy model. Researchers analyzed how parameters such as MIMO configuration, traffic load, and RF power affect energy performance. The findings identify idle power and non-linear power amplifier efficiency as key determinants of network energy use and propose automated ETSI-based test methodologies to guide sustainable O-RAN deployment and energy optimization.
“Energy Efficiency Testing in a Commercial O-RAN System”
This talk extends insights on energy efficiency testing and power modeling of O-RAN radios to real-world operator environments. Through collaboration between the Rutgers/ONF/Aether NOFO-1 Test and Evaluation (T&E) R&D project and the ORCID T&E Lab, the team conducted end-to-end power measurements across a multi-sector, multi-band commercial O-RAN deployment. The results validate the energy test methodology and models. The data confirms that energy efficiency improves when CU/DU overheads are distributed across more RUs and cells, and that RF output power is a strong predictor of O-RU energy consumption. A newly developed multi-band O-RU power model, validated with field data, achieves sub-1% prediction error. The model can be parametrized with simple tests and offers a practical tool for test specification, operator benchmarking, and energy-savings optimization.
These presentations mark a major milestone in the NTIA-funded O-RAN Energy Efficiency Research Program, highlighting collaborative progress toward building sustainable, open, and data-driven RAN infrastructures. Together they demonstrate how rigorous testing, cross-vendor validation, and power modeling can drive actionable strategies for reducing network energy consumption while maintaining performance and interoperability across the O-RAN ecosystem.
