The European project DAEMON, coordinated by the IMDEA Networks Institute since January 2021, has concluded after reaching significant milestones in advancing mobile communications technology. This project has paved the way for effective 6G networks, providing clear benefits to society in terms of faster and ubiquitous communication infrastructures capable of supporting new services.
The main objective of the project was to demonstrate how Artificial Intelligence (AI) can be practically integrated into mobile network architectures at a production level and automate system management beyond 5G and 6G. According to Marco Fiore, a researcher at IMDEA Networks and coordinator of the DAEMON project, “the native AI architectural model proposed by DAEMON has been adopted by the 5G PPP and 6G IA Architecture Working Groups, giving it the potential to be promoted in standardization bodies such as 3GPP.” Additionally, the project has generated two pilots currently in Telefónica’s production network, five solutions part of the European Commission’s Innovation Radar initiative, nine patents, and over 100 scientific articles, which constitute a valuable legacy.
Researchers at IMDEA Networks, particularly the Network Data Science group, have contributed to personalizing loss functions for AI tailored to networks, introducing innovative concepts such as meta-loss learning. They have also demonstrated how AI models can be directly integrated into programmable network hardware to classify traffic flows with delays of just tens of nanoseconds.
Challenges Addressed
The DAEMON project faced various challenges, including:
Understanding the limits of AI for mobile networks: A critical and systematic analysis was carried out to identify which network management tasks can be adequately addressed with AI models, providing solid guidelines for using machine learning in automating next-generation mobile networks.
Developing AI-powered network management solutions: Building on the above guidelines, a range of algorithms tailored to the specific requirements and limitations of mobile network environments was implemented, advancing automation of key management tasks.
Designing an end-to-end AI-native architecture: The first model of AI-native mobile network architecture was outlined, enabling coordinating multiple instances of AI deployed in the network.
Scientific Production and Future Research
The research conducted in DAEMON has made a milestone in its field, with solutions implemented in relevant production network environments. The project has published over 30 articles in high-level conferences and journals, including ACM SIGCOMM, ACM MobiCom, IEEE INFOCOM, ACM/IEEE Transactions on Networking, and IEEE Transactions on Mobile Computing.
While the DAEMON project has concluded, research continues with ORIGAMI, a successor European project initiated in January 2024. The pursuit of mobile networks with lower latency, higher performance, and reliability is advancing, with the promise of improvements in various sectors and social activities.
DAEMON has left a significant mark on the evolution of mobile networks, and the future appears even more promising with projects like ORIGAMI driving innovation forward.
Source: IMdea