SciTransfer
SEASON · Project

AI-Driven Energy Efficient Optical Networks for Beyond 5G Connectivity

digitalTestedTRL 4

Imagine the internet's backbone as a highway. Right now, we are running out of lanes to handle all the new data from VR and AR. This project builds a 'multi-story' highway using special fibers and AI to manage traffic automatically, making it faster and using much less electricity.

By the numbers
16
partners
10
industry partners
62%
industry ratio
The business problem

What needed solving

Current optical networks struggle to scale capacity for beyond 5G services like VR/AR without massive increases in energy consumption and manual management complexity.

The solution

What was built

A sustainable transport network architecture using Multi-Band and Space Division Multiplexing, supported by AI-driven autonomous control systems and power-efficient transceivers.

Audience

Who needs this

Tier-1 Telecom OperatorsOptical Hardware VendorsHyperscale Data Center Providers5G/6G Infrastructure Developers
Business applications

Who can put this to work

Telecommunications
enterprise
Target: National Telecom Operator

If you are a network operator dealing with skyrocketing data traffic from VR/AR services — this project developed a joint Multi-Band and Space Division Multiplexing architecture that increases capacity while reducing energy consumption.

Cloud Computing
enterprise
Target: Data Center Provider

If you are a provider dealing with high costs of optical-electrical-optical conversions — this project developed converged packet-optical solutions and smart NICs that reduce these conversions to save power and cost.

IT Infrastructure
any
Target: Network Equipment Vendor

If you are a vendor dealing with rigid, manual network management — this project developed AI/ML-aided service management and sliceable transceivers that allow for autonomous, near-real-time network operations.

Frequently asked

Quick answers

How does this affect the total cost of network ownership?

Based on available project data, the project targets cost-effective capacity increases by reducing the number of O/E/O conversions and using power-efficient Digital Signal Processing.

Can this be scaled to a national level?

Yes, the architecture is designed to span access, aggregation, and metro/long-haul segments, specifically to support beyond 5G requirements across these layers.

What is the IP and licensing status of the developed technology?

Based on available project data, specific licensing terms are not mentioned, but the project involves 10 industrial partners including major vendors like Ericsson and Infinera.

How does this integrate with existing 5G hardware?

It integrates via open packet/optical white boxes using smart Network Interface Cards (NICs) and Data Processing Units (DPUs) to decouple control from management.

What is the timeline for deployment?

The project runs from 2023-01-01 to 2025-12-31, with the first period focused on defining requirements and modeling data plane technologies.

Consortium

Who built it

The project is heavily industry-led with a 62% industry ratio, comprising 10 companies including tier-1 operators (Telefonica, TIM) and global vendors (Ericsson, Infinera, ADVA). This high level of industrial involvement, combined with 3 SMEs and 6 academic/research centers, suggests a strong path toward commercialization and practical market application.

How to reach the team

Contact CONSORZIO NAZIONALE INTERUNIVERSITARIO PER LE TELECOMUNICAZIONI

Next steps

Talk to the team behind this work.

Contact us to connect with the SEASON consortium for early adoption of sustainable optical networking.