Project Abstract
5G/6G systems are well known to go beyond the classical “one-fits-all” paradigm of previous radio mobile network generations.
They provide not only the possibility to dynamically create, update and delete custom “private networks” in the form of network slices for supporting special applications with challenging heterogeneous needs, but even to split and to distribute the overall system across various stakeholders and domains.
These stakeholders can span from infrastructure or connectivity providers to vertical user/application providers.
To support this new paradigm, 5G/6G systems are composed of an ever increasing number of control and management components acting on the different domains, which are able to expose and consume 5G and edge computing resources “as-a-Service.”
The inner complexity of this multi-tenant and multi-domain environment is envisaged to be handled by the use of “Intent-Based” APIs, able to abstract and to separate diverse Artificial Intelligence (AI) engines providing automated operations and the
reinforcement/optimization of policies for each Stakeholder domain.
In this complex and radically new environment, the cascade effect of a change in the optimization policy (or even of a single reconfiguration of resources) by a Stakeholder can be hardly mapped onto the effect produced on the overall ecosystem. In detail, this can arise especially in those cases where AI engines of Stakeholders have partially conflicting policies and objectives, and might trigger potential network instability or performance decay.
The 6GTWINS project aims to address this issue by exploiting the concept of Digital Twins and applying it to network automation and orchestration. In particular, the project will design solutions for exploiting Digital Twins for What-If analysis and for speeding up AI/ML training.
Obviously, the project is not so ambitious to completely fill the research gap in this complex and multi-facet problem, but to move some first steps towards novel approaches and promising technologies.
In fact, the project decided to restrict the application scenario of the studied technologies to a representative use-case, namely a Digital Twin-based network orchestration framework for 5G/6G slicing.
Achieved results
The main achievement results are presented publicly in the form of a Virtual 6GTWINS Workshop on YouTube.
The workshop is accessible by the link at the bottom of the page, while the individual presentations are available in the following videos.
Publications
1. R. Bolla, R. Bruschi, A. Gallo, C. Lombardo and N. S. Martinelli, "To Scale or Not to Scale? Understand the Overhead of Container Scaling Operations," 21st Internat. Conf. on Distributed Computing in Smart Systems and the Internet of Things (DCOSS-IoT), Lucca, Italy, 2025, pp. 890-894, doi: 10.1109/DCOSS-IoT65416.2025.00135.
2. M. Akbari, R. Bolla, R. Bruschi, C. Lombardo, N. S. Martinelli and B. Siccardi, "Observe to Sustain – How to Enable Beyond 5G Networks to Target Sustainability Goals," 35th IEEE Internat. Symp. on Personal, Indoor and Mobile Radio Comm. (PIMRC), Valencia, Spain, 2024, pp. 1-7, doi: 10.1109/PIMRC59610.2024.10817341.
3. A. Caruso, C. Grasso, R. Raftopoulos, G. Schembra, "IDLE: A Digital Twin Framework for 6G Network Intelligence," IEEE CAMAD 2024, Athens, Greece, October 21-23, 2024.
4. A. Caruso, C. Grasso, R. Raftopoulos, G. Schembra, “FALCON: FANET-Aware Learning and digital twin CONtrol framework,” Volume 251, 2026, 108481, ISSN 0140-3664. https://doi.org/10.1016/j.comcom.2026.108481
5. Tailai Song, Paolo Garza, Michela Meo and Maurizio M. Munafò, "Towards the Detection of Unobservable Losses in Real-Time Communications," 2024 IEEE 30th International Symposium on Local and Metropolitan Area Networks (LANMAN), Boston, MA, USA, 2024, pp. 21-26.
6. Andrea Marin, Marco Ajmone Marsan, Michela Meo, Matteo Sereno, “Queuing models of links carrying streaming and elastic services,” Computer Networks, Volume 244, 2024,ISSN 1389-1286.
7. POLVERINI, Marco, et al. Avoiding SDN Application Conflicts With Digital Twins: Design, Models and Proof of Concept. IEEE Transactions on Network and Service Management, 2026, 23: 2038-2050.
8. POLVERINI, Marco, et al. Guiding Network Function Virtualization Orchestration Through the Digital Twin Technology. In: 2025 IEEE 11th International Conference on Network Softwarization (NetSoft). IEEE, 2025. p. 1-6.
9. J. Sengendo, F. Granelli, “Building Network Digital Twins Part I: State Synchronization,” 2024 3rd International Conference on 6G Networking (6GNet), Paris, France, 2024, pp. 182-188, doi: 10.1109/6GNet63182.2024.10765759
10. J. Sengendo, F. Granelli, “Optimizing Network Traffic Prediction for Network Digital Twins: The Impact of Look-Back Period and Forecast Horizon,” 2025 IEEE Wireless Communications and Networking Conference (IEEE WCNC 2025).
11. D. D. Pietra, K. Govindarajan, F. Granelli, A. Rosani and N. Garau, "Drone Agents: learning to fly to learn how to see," 2025 IEEE International Conference on Advanced Visual and Signal-Based Systems (AVSS), Tainan, Taiwan, 2025, pp. 1-6, doi: 10.1109/AVSS65446.2025.11149899
12. M. Devigili et al., "ML-Based Modeling of EDFA Pluggable Modules for OSNR Estimation," 2025 International Conference on Optical Network Design and Modeling (ONDM), Pisa, Italy, 2025, pp. 1-6, doi: 10.23919/ONDM65745.2025.11029334
13. M. Chiarani, S. Roy, C. Verikoukis and F. Granelli, "XAI-Driven Client Selection for Federated Learning in Scalable 6G Network Slicing," ICC 2025 - IEEE International Conference on Communications, Montreal, QC, Canada, 2025, pp. 1881-1887, doi: 10.1109/ICC52391.2025.11161532
GitHub repositories
https://github.com/fabrizio-granelli/Amarisoft.digital.twin
https://github.com/fabrizio-granelli/Amarisoft-5G-Digital-Twin-v2
