Federated AI and reactivity of network alert systems

Orange’s 2021 Research Exhibition is an opportunity to learn about the French National Research Agency’s project “Discret”, a joint innovation by Orange, Gustave Eiffel University and Troyes University of Technology.

“With federated AI, mobile network usage data is processed in edge mode, i.e. as close as possible to the aerial.”

 

With federated AI, mobile network usage data is processed in edge mode, i.e. as close as possible to the aerial.

This makes it a technology that has significant potential to improve the reactivity of network alert systems. Unlike centralised AI, sensitive data is protected, since it does not circulate over the network. Calculations, performed locally, bring results more quickly. The models calculated during the learning phase are shared between relays. Lastly, without the need to transmit data and cool a central server, the energy efficiency of the process is improved.

Read also on Hello Future

Two people are seated in front of a computer, discussing a project. Spools of thread are visible on the table.

How to avoid replicating bias and human error in LLMs

Discover

Combining Clustering and AI for Congestion-Free Mobile Networks

Discover

NORIA: Network anomaly detection using knowledge graphs

Discover
GettyImages - machine learning - research recherche

Machine-Learning-Based Early Decision-Making (ML-EDM)

Discover

Green and Local Energy Exchanges and Optimizing Network Consumption

Discover
A 5GCroco - Stellantis test car

Connected Cars En Route to Autonomy

Discover

Encrypting or classifying Internet traffic: do we have to choose?

Discover

Advocating for ethical and responsible AI by design

Discover