ANIARA – the joint European initiative to shape an intelligent and sustainable 6G platform

The AI-NET-ANIARA project contributes to European leadership in communication technology

AI-NET-ANIARA, CELTIC NEXT's flagship research project, unveils an extensive list of compute and AI technological enablers after successful proof-of-concept assessments, paving the way for 6G development. Three EU countries and 20 partners were involved in the project from Sweden, Germany, and the UK. The project goal was to complement the evolution of 5G with crucial technical enablers towards an intelligent and sustainable 6G platform, offering services beyond pure communication, including compute & AI.

In Figure 1, proposed by the International Telecommunication Union (ITU), the blue hexagon represents the envisioned expansion of network usage scenarios towards 2030, which is considered the 6G timeframe by the telecommunications industry at large. This shows the evolution of 5G communication services and three new corners that will be required to meet the needs of digital societies beyond 2030: Ubiquitous (global) connectivity, integrated sensing capabilities, and integrated compute and AI capabilities.

Realizing this ambitious vision requires starting early and working together with both industry and the research community. The Celtic-Next AI-NET-ANIARA project provided a good community for exploring this goal. The ANIARA project goal (quoting the project proposal) "accelerating digital transformation by the efficient use of a highly integrated and flexible edge infrastructure that is programmable across all its components, from the basic connectivity setup to fully virtualized network functions and application components" turned out to be spot on to complement the evolution of 5G with crucial technical enablers towards an intelligent and sustainable 6G platform. The 6G aspects and usage scenarios where the ANIARA project contributed the most are indicated in Figure 1.

Fig 1: 6G aspects and usage scenarios. Source: "Recommendation ITU-R M.2160-0: Framework and overall objectives of the future development of IMT for 2030 and beyond."

AI-NET-ANIARA main contributions

The ANIARA project develops compute and AI enablers for future mobile networks. In Figure 2, we provide a simplified view of a mobile network with its sites and functional layers. The ANIARA project contributed to the areas highlighted in blue boxes, and the thin blue line frames the focus of these contributions on network edge sites.

Fig 2: Simplified view of mobile network

Starting from the bottom up, the project contributed to a sustainable and intelligent 6G network platform with the following:

1. Energy-efficient edge infrastructure at access sites;
2. Automated management of the distributed platform, which includes both resource orchestration for distributed edge resources as well as intelligence support for 6G network functions; and
3. 6G platform services on the edge exposed for society, enterprise, and industrial applications.

While the enablers on these layers are applicable to many emerging use-cases, we chose smart city and smart manufacturing applications as two highlighted use-cases to exemplify the value proposition and usability of the proposed technology innovations.

Achieved results

Below is a summary of achieved results from the ANIARA project:

• Carrier grade AI technologies for telecom edge automation, and intelligent management in support of services with guarantees on performance and energy consumption.
• Novel AI concepts such as robust and reliable federated learning, reinforcement learning for service mesh, intelligent feature selection, and transfer learning.
• MLOps infrastructure for AI. ANIARA extended the Feature store solution by ANIARA partner Hopsworks to support real-time model serving all the way to the edge.
• Execution runtime to process edge applications using WebAssembly. A dynamic computational offloading solution was developed and its potential as a future 6G service successfully demonstrated on the smart city application by ANIARA partner UNIVRSES.
• A modified Kubernetes scheduler enabling the allocation of workloads based on the energy state of nodes and devices. This modification improved the energy efficiency of the system. Furthermore, the project designed a new resource scheduling mechanism for Kubernetes that can closely match the time-varying traffic profile of users. This led to a 30% improvement in resource efficiency.
• Designed, built, and operated two versions of ANIARA edge data center demonstrators. These demonstrators include a fresh air-cooling approach, where the cooling system is highly integrated in the Edge node design. An intelligent power system ensures that the servers receive enough power when external power supply is limited and incorporates a photovoltaic system to reduce the amount of power from the grid, improve sustainability, and enhance robustness.
• Developed new edge processing devices deployed in Stellantis manufacturing sites that monitor energy consumption and machine condition. This data is then transmitted wirelessly to optimize the energy consumption in the paint shop.

Knowledge sharing and collaboration part of the project impact

The ANIARA project organized a set of dissemination activities such as dedicated workshops with small and medium enterprises, as well as workshops and demonstrations at major conferences to share project insights beyond the project partner organizations. Several talks were given at selected industrial and scientific venues, which can potentially impact the long- and short-term development plans of the relevant industry domains.
We also established collaboration with different national initiatives to enhance the reach and impact of the project results and promoted AI & compute enablers developed by ANIARA.

We are proud that AI-NET-ANIARA successfully delievered on its project promises of accelerating digital transformation by the efficient use of a highly integrated and flexible edge infrastructure that is programmable across all its components.

The AI-NET-ANIARA officially concluded on April 30, 2024. We continue the work in Ericsson to develop the outcome of the project for possible integration into products.

The authors would like to thank Ericsson colleagues Peter Öhlén, Andreas Johnsson, Johan Sjöberg, and Björn Skubic for their contributions and support.

Learn more

Scientific paper describing the project goals: “ANIARA Project-Automation of Network Edge Infrastructure and Applications with Artificial Intelligence”, ACM SIGAda Ada Letters 42, no. 2 (2023) by John, Wolfgang, Ali Balador, Jalil Taghia, Andreas Johnsson, Johan Sjöberg, Ian Marsh, Jonas Gustafsson et al. Read the abstract

Paper: "On Heterogeneous Transfer Learning for Improved Network Service Performance Prediction," 2021 IEEE Global Communications Conference (GLOBECOM), Madrid, Spain, 2021, pp. 1-6, by F. G. Sanz, M. Ebrahimi and A. Johnsson. Read the abstract

Paper: "Policy-Induced Unsupervised Feature Selection: A Networking Case Study," IEEE INFOCOM 2022 - IEEE Conference on Computer Communications, London, United Kingdom, 2022, pp. 750-759, by J. Taghia, F. Moradi, H. Larsson, X. Lan, M. Ebrahimi and A. Johnsson. Read the abstract.