Although there is not a unique understanding of what 6G will be, several initiatives are ongoing that have put forward highly advanced visions of potential concepts and preliminary technologies that will form 6G networks. With the current deployment of 5G networks, high data rate and low latency are provided for communication, in addition to some first steps towards deployments that also provide commercial localization services. 6G mobile networks, however, will go far beyond the use cases that can be covered by 5G, enabling not only significantly improved network performance but also substantially more complex services that rely on location and context information gathered by the network. In particular, 6G mobile networks will enable orders of magnitude higher localization accuracy and lower latency than prior technologies. This will be a unique opportunity to design new services and analytics, but also a threat for privacy. For this reason, this project will design native privacy-preserving machine learning mechanisms for 6G networks in order to manage the massive amount of data generated by services in 6G networks, based on emerging Federated Learning techniques. The final demonstrator will integrate the developed modules within the mobile network and will be demonstrated using testbeds comprising data servers, edge nodes and end-user devices.

Funded by the Ministry of Digital Transformation and Civil Service and the European Union-NextGenerationEU through the UNICO-5G R&D Program of the Spanish Recovery, Transformation and Resilience Plan.