SHINE - Smart High-Performance and Secure Optical Access and Transport Networks for sustainable 6G and AI

Ico_CTTC
Start: 01/09/2025
End: 31/08/2028
Funding: National
Status: On going
Research unit:
Packet Optical Networks and Services (PONS)
Acronym: SHINE
Call ID: PID2024
Code: PID2024-159781OB-I00

AI and 6G will reshape society and the economy, but they impose new demands on fixed optical access and transport networks. Generative AI drives rising data-center interconnect (DCI) traffic, requiring automated, energyefficient management. 6G, targeting THzband links, will centralize RAN processing in datacenters, pushing highcapacity, lowlatency fronthaul. Networks must adopt quantumsafe security. 

SHINE reimagines fixed networks for 6G and AI by coupling distributed edge/core datacenters with onehop optical paths from users to compute. By minimizing intermediate packet handling, SHINE targets high capacity, low latency, energy efficiency and security across multifiber/multiband infrastructures. It leverages multicarrier multiplexing, PON, WDM, Band Division Multiplexing (BDM) and Spatial Division Multiplexing (SDM), and integrates quantumsafe protections. 

Key objectives: 

  1. Sustainable capacity scaling: design and validate multiband over SDM transmission/switching, integrating WDM+SDM for capacity, resource optimization, and programmable control. 
  1. Service continuity & reliability: build adaptive closedloop AI/ML for transmission, monitoring, fiber sensing, and Digital Twins for prediction/automation. 
  1. Optical access: advance softwaredefined access, optical bypass, and programmable spatially diverse P2MP for seamless segment integration. 
  1. Edge & AIdriven programmability: enhance TelcoCloud with SmartNICs/GPUs/TPUs to enable realtime observability, innetwork processing, and scalable inter/intraDC integration.
  1. Security & quantum integration: explore QKD, converge classical/quantum networks, automate security, and use blockchain for resilience. 
  1. Sustainability & energy efficiency: create energyaware models, optimize power, and apply AIdriven allocation to balance throughput vs. energy. 
  1. Prototype & demonstration: validate on the ADRENALINE testbed with KPIs and configurations for proofofconcept. 
Raül Muñoz
PI/Project Leader
Michela Svaluto Moreolo
Researcher
Ramon Casellas
Researcher
Ricardo Martínez
Researcher
Luca Vettori
Researcher
Lluis Gifre
Researcher
Pol Alemany
Researcher
Josep M. Fàbrega
Researcher
Laia Nadal
Researcher
Francisco Javier Vilchez
Researcher
Waleed Akbar
Researcher
No results found