Network intelligence for aDAptive and sElf-Learning MObile Networks

Projektinformationen

DAEMON

ID Finanzhilfevereinbarung: 101017109

DOI

10.3030/101017109

Projekt abgeschlossen

EK-Unterschriftsdatum 8 Dezember 2020

Startdatum 1 Januar 2021

Enddatum 31 März 2024

Finanziert unter

• INDUSTRIAL LEADERSHIP - Leadership in enabling and industrial technologies - Information and Communication Technologies (ICT)

Gesamtkosten

€ 4 997 025,00

EU-Beitrag

€ 4 997 025,00

Koordiniert durch

FUNDACION IMDEA NETWORKS
Spain

Periodic Reporting for period 2 - DAEMON (Network intelligence for aDAptive and sElf-Learning MObile Networks)

Berichtszeitraum: 2022-01-01 bis 2024-03-31

Zusammenfassung vom Kontext und den Gesamtzielen des Projekts

DAEMON is a 39-month Research and Innovation Action (RIA) funded by the European Union H2020 framework programme under call H2020-ICT-2018-20. The project has studied solutions that help Beyond-5G (B5G) mobile networks meet high expectations in terms of support for very diverse services, near-zero latency, apparent infinite capacity, and 100% reliability and availability. The main goal of DAEMON was developing and implementing innovative and pragmatic approaches to the design of the Network Intelligence (NI) that orchestrate physical resources and Virtual Network Functions (VNFs) in next-generation mobile systems. Notably, DAEMON has departed from the current hype around Artificial Intelligence (AI) as the silver bullet for any mobile network management task and realize a systematic integration of varied Machine Learning (ML) models into network architectures that ensures an effective operation of NI in B5G infrastructures.
The project activities have been structured around the following set of technical objectives.
Objective 1 – Designing a “NI-native architecture” for B5G systems. DAEMON aimed at proposing updates to the network architecture so that it natively supports NI operations. This objective stands on two pillars: (i) enhancing the current architectural vision in B5G standardization efforts so that it implements a well-structured support for coordinated NI across domains, and (ii) fostering a tighter incorporation of NI into the network infrastructure, bringing it closer to the end user and hardware.
Objective 2 – Developing specialized NI-assisted network functionalities for B5G systems. DAEMON designated a concrete list of eight key network functionalities that span a range of operation timescales, network planes and domains. For all these functionalities, the project devised and implemented NI algorithms able of taking advantage of the proposed NI-native architecture, populating the proposed NI-native architecture with actual intelligence.
Objective 3 – Establishing fundamental guidelines for a pragmatic design of NI. DAEMON innovated the way NI is conceived, by leveraging emerging trends in AI and re-thinking how AI is applied to the network infrastructure environment. To this end, DAEMON (i) drafted guidelines on the most appropriate ML tools for specific networking tasks, (ii) designed loss functions for AI model training that are tailored to the network management goal, and (iii) developed AI models that can adaptively trade off accuracy for network-relevant metrics like latency or complexity.

Arbeit, die ab Beginn des Projekts bis zum Ende des durch den Bericht erfassten Berichtszeitraums geleistet wurde, und die wichtigsten bis dahin erzielten Ergebnisse

The work carried out in the project has been structured in four technical Work Packages (WPs).
In WP2, activities have focused on four main tasks: (i) drafting functional and performance requirements from the target NI-assisted functionalities; (ii) outlining NI orchestrator and NI interfaces to integrate NI natively in B5G systems; (iii) identifying relevant AI techniques for concrete networking problems and understanding their limitations; (iv) developing methodologies for novel AI-based approaches that are tailored to real-world NI problems.
In WP3, the work has focused on two main aspects: (i) drafting a functional architecture for real-time NI; (ii) designing specific algorithms that populates the architecture above.
In WP4, four parallel and intertwined research threads are pursued: (i) studying energy-aware VNF placements; (ii) designing capacity forecasting models to assisting NI functionalities; (iii) developing a clean-slate automated anomaly detection approach targeting large-scale IoT networks; (iv) developing zero-touch solutions for resource allocation decisions in MANO systems.
In WP5, the following tasks have been carried out: (i) identifying the main simulation/emulation tools that are used in the project in order to validate the NI solutions; (ii) identifying the available sites/testbeds that are available in the project and can be used for the prototyping activities and for providing the project demos; (iii) identifying the available datasets that can be used during the validation process; and, (iv) leveraging all tools above to perform a dependable evaluation and demonstration of the solutions developed in WP3 and WP3.

Fortschritte, die über den aktuellen Stand der Technik hinausgehen und voraussichtliche potenzielle Auswirkungen (einschließlich der bis dato erzielten sozioökonomischen Auswirkungen und weiter gefassten gesellschaftlichen Auswirkungen des Projekts)

By the end of its execution, DAEMON has achieved and often exceeded its targets, as follows.
At the architectural level, DAEMON has produced a list of requirements for a NI-native B5G network architecture, including a complete design and initial implementation of a NI Plane (NIP) capable of managing the lifecycle of AI models deployed across network micro-domains, of handling their conflicts, and of leveraging their synergies. In this context, the project has introduced a formal way to represent individual NI instances in the NIP via an original N-MAPE-K approach. The NIP concept has evolved through a series of consortium-wide papers and has been adopted by 5G PPP and later 6G IA as the basis to develop a NI Stratum in their influential architectural model for 6G systems.
Concerning the design of NI algorithms, DAEMON has produced a list of requirements for the AI that shall empower eight key network management tasks, as well as lists of (i) concrete recommendations on the limits of AI models for NI operation and (ii) general guidelines on how to tailor AI models to the unique specificity of network environments. Based on these requirements, recommendations and guidelines, the project has designed, implemented, and validated in dependable settings a wide range of NI solutions targeting those same eight tasks. The solutions have led to two pilots (TRL 5) and a number of demonstrators that were showcased at venues such as Mobile World Congress and received recognitions such as the IEEE NetSoft Best Demo Award.
The project was highly successful in terms of scientific impact, with over 100 published papers at peer-reviewed international conferences and journals, around one third of which in top-tier (i.e. JCR Q1 or A/A* CORE ranking) venues. DAEMON results were presented in 11 times in the main program of the IEEE INFOCOM conference, and appeared at ACM SIGCOMM, ACM MobiCom, ACM SIGMETRICS, ACM CoNEXT, AAAI, and a range of IEEE/ACM Transactions journals.
The impact of DAEMON was also substantial on (i) standardization, with 45 contributions submitted to main bodies like O-RAN, ETSI, 3GPP and ITU, (ii) patenting, with 9 patents filed that originated from the project activities, (iii) innovation, with 5 solutions developed in the project accepted in the Innovation Radar of the European Commission, (iv) open-sourcing, with 8 contributions to large initiatives like srsRAN and Eclipse Zenoh plus 20 minor open assets, (v) workshops organization, with 2 industry events and 7 scientific events coordinated.

DAEMON Architecture