NVS2 – 6G & Open network architectures
Thursday, 5 June 2025, 16:00-17:30, room 1.C
Session Chair: Tao Chen (VTT Technical Research Centre of Finland LTD, FI)
A Framework for Generic Energy Efficiency Metrics: Defining Ideal Linear Reference Systems
Tobias Hoßfeld (University of Würzburg, Germany)
Energy consumption (EC) and energy efficiency (EE) are crucial for operating communication networks and are key concerns for upcoming 6G networks. Energy intensity metrics or bit-per-Joule metrics are increasingly used to estimate the energy costs and benefits of changes in data volumes and to quantify EE. While these metrics typically integrate annual energy consumption and data transmission, they can be less reliable at shorter time scales due to weak correlations between transmitted data and energy consumption, sometimes leading to inaccurate conclusions. Recently, the generic EE measures consumption-related EE (CrEE) and output-related EE (OrEE) were suggested, which come with some advantages over existing EE measures. Thereby, efficiency is expressed by comparing the EC of a real system depending on the traffic load with that of an ideal system. However, if the ideal system is not known, the question arise how to construct the ideal system function, which is the EC depending on the traffic load. As a key contribution, we provide a practical guideline to construct such ideal systems by utilizing desired or known properties of the ideal system.
AI-Native 6G Networks: the 6GARROW Integrated Device-Network Approach
Emilio Calvanese Strinati (CEA-LETI, France); Nicolas Cassiau (CEA-Leti, Université Grenoble Alpes, France); Chan-Byoung Chae (Yonsei University, Korea (South)); Pierre Dal Zotto (Grenoble Ecole de Management, France); Nicola di Pietro (Hewlett Packard Enterprise, Italy); Louis-Adrien Dufrène (Orange Research, France); Thomas Haustein (Fraunhofer Institute for Telecommunications, Heinrich-Hertz-Institut, Germany); Riku Jäntti (Aalto University, Finland); Joongheon Kim (Korea University, Korea (South)); Seong-Lyun Kim (Yonsei University, Korea (South)); Tae-Yeon Kim (ETRI, Korea (South)); Tanesh Kumar (Aalto University, Finland); Quentin Lampin and Guillaume Larue (Orange Research, France); Matti Latva-aho (University of Oulu, Finland); Jemin Lee (Yonsei University, Korea (South)); Petri Mähönen (Aalto University, Finland); Dileepa Madhubhashana Marasinghe (University of Oulu, Finland); Davide Montagno Bozzone (HPE, Italy); Markus Dominik Mueck (Intel Deutschland GmbH, Germany); Jeonghun Park (Yonsei University, Korea (South)); Premanandana Rajatheva (University of Oulu, Finland); Zoran Utkovski (Fraunhofer HHI, Germany)
This paper introduces the concept of AI-native radio access networks (RANs) and integrated device-network approaches, which aim to seamlessly embed artificial intelligence (AI) capabilities across devices (e.g., smartphones, IoT sensors, autonomous vehicles) and network infrastructures. By embedding AI deeply within RANs and fostering tight integration between devices and network architectures, we unlock unprecedented levels of intelligence, efficiency, flexibility, and performance in wireless communication systems. This integration enables devices and networks to collaboratively process, adapt, and optimize data in real-time, creating a synergistic AI-enabled ecosystem that forms the backbone of AI-Native 6G. Unlike traditional wireless communication systems, which rely on static configurations and manual interventions, AI-native networks dynamically adapt to changing conditions, autonomously optimize resource allocation, and respond to diverse user demands and environmental factors. The 6GARROW project explores these transformative approaches, demonstrating how AI-native RANs and integrated device-network solutions deliver higher throughput, lower latency, and improved reliability. Beyond enhanced performance, this paradigm shift supports a diverse range of innovative services, including spatio-temporal communications, critical and compute-AI applications, omnipresent IoT, immersive user experiences, and global broadband connectivity. This paper outlines the vision, methodologies, and potential impact of the 6GARROW project, highlighting its role in advancing toward a more connected, intelligent, and adaptive wireless ecosystem for the future.
Research and Innovation in Europe on Cloud for 6G Networks
Toon Norp and Maria Raftopoulou (TNO, The Netherlands); Pierre-Yves Danet (6G-IA, Belgium); Claudio De Majo (Trust-IT Services, Italy); Prachi Sachdeva (TNO, The Netherlands); Konstantinos Trichias (6G Smart Networks & Services Industry Association (6G SNS IA), Greece & National Technical University of Athens (NTUA), Greece); Pooja Mohnani (Eurescom, European Union)
Cloud is becoming increasingly important in the telecommunication sector in view of 6G networks. Moreover, cloud has gained major attention after the European Commission published a white paper proposing the creation of the “3C Network”, an European telco edge/cloud infrastructure, which aims to increase collaboration between European players and ensure innovation, and economic and digital security in Europe. In a 6G network, a telco cloud can play three different roles. First it is a basis to implement 6G Radio Access Network (RAN) and other network functions. Second the telco cloud can implement Artificial Intelligence (AI) and other applications for 3rd party providers in the edge. Third it allows devices/cars/drones/persons to outsource compute capabilities to the 6G edge. Currently, there are various ongoing European activities that address cloud, but they often have different focus and requirements. This paper provides an overview of ongoing activities related to telco edge/cloud. The analysis on the ongoing activities showed a clear trend towards providing open-source and/or standardised solutions. Furthermore, it has been identified that the topic of security is widely addressed.
DMMP: the AI Service Platform with O-RAN Toward AI-RAN
Po Chun Hung, Yu Kai Lin, Yu-Chiao Jhuang, Tuck-Wai Choong and Jenq-Shiou Leu (National Taiwan University of Science and Technology, Taiwan)
As 5G commercialization and 6G research accelerate, high-frequency bands are becoming the mainstream spectrum for future communications, emphasizing the critical role of AI in network optimization and decision-making. This study proposes the Distributed MLOps Management Platform (DMMP) to integrate MLOps into the 6G network system, enabling automated model deployment, monitoring, and optimization while exploring its feasibility with O-RAN.
Leveraging a subscription-based architecture, DMMP interacts with third-party AI/ML platforms to support power forecasting and network function scheduling in Taiwan, demonstrating high scalability and dynamic orchestration. Its core features include scalable edge model deployment and version management, enhancing AI-RAN’s real-time scheduling and stability in dynamic environments. By integrating with O-RAN, DMMP further strengthens AI-RAN capabilities, laying a robust foundation for the communications industry as it advances into the 6G era.
An Approach for Key Value Indicator Evaluation: Enabling Consistent Assessment of Societal Impacts
Hassan Osman (Real Wireless, United Kingdom (Great Britain)); Julie Bradford (Real Wireless Limited, United Kingdom (Great Britain)); Stuart Mitchell (Real Wireless, United Kingdom (Great Britain)); Claudia Chiavarino (Istituto Universitario Salesiano Torino)
This paper presents a harmonized methodology for evaluating Key Value Indicators (KVIs) to address the challenges of inconsistent assessments across projects. The proposed approach provides a framework for analyzing societal impacts, offering insights beyond traditional business case analysis or environmental impact studies. By enabling meaningful comparisons between technologies in terms of their benefits, this methodology supports informed decision-making in both business and policy contexts, particularly in public-private partnerships. The framework’s adaptability also allows for its application across diverse use cases, making it a valuable tool for fostering innovation and driving impactful outcomes. Addressing societal impact poses significant challenges due to its inherently qualitative nature, making it difficult to quantify compared to environmental and business dimensions. This study focuses specifically on assessing societal progress by adopting a relative measurement approach. By having baseline data points, we aim to generate incremental metrics that reflect improvements over time. This method provides a systematic way to track progress within a broader context, offering insights into trends and changes in societal outcomes.
