B5G/6G inaugurates the era of massive and extremely heterogeneous services that target several vertical industries involving both human and machines. This raises new challenges regarding the sustainability of these next-generation networks. In this respect, Artificial intelligence (AI) techniques such as deep, federated and reinforcement learning pave the way to achieve cost-efficient and green B5G/6G networks.
Driven by the massive availability of network performance and configuration data, AI techniques have great potential to analyze patterns and take fast decisions. This will help to achieve the promise of automation and sustainability in B5G/6G networks by increasing energy-efficiency and lowering operational expenditures (OPEX), as well as management complexity. Specifically, the data-driven trend is moving from big and centralized data towards distributed and small data, bringing the analysis and decision closer to the data collection points in the network which might reduce overhead and the underlying costs. This calls for the use of new AI techniques, going beyond traditional ones, in order to optimize B5G/6G networks functions in all technological domains, i.e., RAN, edge, Cloud and Core.
Therefore, AI is a key component to achieve a sustainable B5G/6G ecosystem, especially by leveraging distributed monitoring, analysis, and decision, which urges to conduct advanced works in the area of AI, either from an architectural or algorithmic point of views.
This workshop aims at bringing researchers together to discuss the opportunities and challenges in the research, design, and engineering of how AI techniques can impact B5G/6G networks design, especially advanced techniques of machine learning. This event will focus on, but will not be limited to, the following subjects of interest:
- Energy-efficient AI enabled B5G/6G network management.
- Energy-efficient AI enabled resources provision and orchestration.
- AI for low complexity in B5G/6G.
- Decentralized AI for sustainable B5G/6G.
- Green communication in AI-enabled B5G/6G.
- Edge learning for sustainable B5G/6G.
- AI-empowered mobile edge computing for B5G/6G.
- Learning-based Edge Computing Services.
- AI-enabled sustainable network slicing of B5G/6G networks.
- Energy efficiency of deploying machine learning over B5G/6G networks.
- AI-empowered dynamic function splitting in Green B5G/6G networks.
- Technologies, concepts, and theories for integrated AI and Green B5G/6G networks.
Adlen Ksentini, EURECOM, France
Mehdi Bennis, University of Oulu, FINLAND
Dr. Bouziane Brik, Bourgogne University, France (firstname.lastname@example.org)
Dr. Hatim Chergui, CTTC, Barcelona, Spain (email@example.com)
Pr. Abderrahmane Lakas, United Arab Emirates University (firstname.lastname@example.org)
Dr. Elli Kartsakli, Barcelona Supercomputing Center, Spain (email@example.com)
Dr. Hassnaa Moustafa, Intel, USA (firstname.lastname@example.org)
Paper Submission Deadline: August 14th, 2021 (FIRM)
Notification of Paper Result: September 15, 2021
Camera-Ready: October 8, 2021
Authors are encouraged to submit full papers describing original, complete research, not currently under review by another conference or journal.
Accepted papers will appear in the workshop proceedings, and be published in the IEEEXplore Digital Library.
All submissions should be done via EDAS
ASNet 2021 is organized with the support of the Horizon 2020 ICT-20 project MonB5G (Grant No. 871780 )