Förbättrad säkerhet och integritet för trådlöst federerat lärande

The main goal of this research is to come up with efficient solutions to mitigate security and privacy threats in wireless federated learning. We aim to develop solutions in the physical layer of the wireless communication system that can provide improvements in security and privacy. Inspired by ideas in the physical layer security literature, we aim to investigate whether the inherent randomness due to channel fluctuations, distortions, and noise introduced by the transceiver hardware can be exploited with carefully designed transmission schemes/protocols to make federated learning robust to privacy and security threats.

Startdatum 2020-08-01
Slutdatum Projektet är avslutat: 2021-01-31

Sidansvarig Publicerad: to 14 jan 2021.