GASTON

The project

GASTON – Green Autonomic and Resilient Optical Multi-Core Networks is a PRIN 2022 PNRR project funded by the Italian Ministry of University and Research. Vital societal functions increasingly rely on digital network infrastructures, and 5G/6G services require optical networks to scale both capacity and resilience. Spatial division multiplexing over multi-core fibres (MCFs) is emerging as the answer to the optical capacity crunch, but it also raises network complexity and poses new survivability challenges. GASTON addresses two of them: reconciling resilience with energy efficiency, and enabling cooperation among vendors, integrators and operators while preserving privacy.

The project delivers architectural and algorithmic solutions for survivable, energy-aware multi-core optical networks (MCONs), machine-learning models for autonomic failure detection and localisation, and federated-learning frameworks for privacy-preserving, multi-stakeholder fault management. All solutions are validated experimentally on the world's first field-deployed multi-core fibre testbed, located in L'Aquila.

Our role

The University of L'Aquila unit, led by Andrea Marotta, leads the work on new strategies for resilient MCONs and on field experimentation, as well as dissemination and data management, and contributes to the federated-learning activity. Its central asset is the L'Aquila field-deployed SDM fibre infrastructure (coupled-core and uncoupled 4-core, uncoupled 8-core, 15-mode and single-mode fibres in a 7 km underground tunnel, part of the CNIT FIBERS national laboratory), which the unit turned into a failure-emulation testbed: a multi-span weakly-coupled 4-core link (25.6 km) with crosstalk-injection stages, a custom 8-channel C-band EDFA box, and a coupled-core 4-core link (69.3 + 6.3 km) with mode-dependent-loss emulation.

NSLab designed the failure scenarios, ran the measurement campaigns (32,000 data points on the weakly-coupled link; 5,400 baseline plus a 24,000-point MDL dataset on the coupled-core link, with 4/16/64-QAM) and shared the quality-of-transmission datasets (BER, Q-factor, OSNR, GMI) with Politecnico di Milano, enabling ML-based failure localisation with up to 96% accuracy. The unit also contributed energy-efficient PON architectures based on dynamic spatial aggregation with NETCONF control, presented at ONDM 2024, OECC 2025 and ICTON 2025.

Facts

• Programme: PRIN 2022 PNRR, MUR, PNRR Mission 4 Component 2 Investment 1.1, funded by the European Union – NextGenerationEU (project P20223LHTR, ERC sector PE7).
• Duration: 24 months, December 2023 – November 2025.
• Coordinator: Prof. Massimo Tornatore, Politecnico di Milano. Research units: Politecnico di Milano, University of L'Aquila. International collaborators: NICT (Japan), UC Davis (USA), SM Optics.
• Budget: MUR grant € 239,985, of which University of L'Aquila € 109,560.
• Unit lead: Andrea Marotta.