FadePrint - Satellite Spoofing Detection via Fading Fingerprinting.

While various methods exist to implement message authentication in different communication layers, the physical layer offers some unique and beneficial features for this purpose. Existing solutions authenticate transmitters at the physical layer by merging deep learning with physical-layer attributes, protecting against impersonation attacks. This approach requires a lengthy and resource-intensive training phase for every new transmitter that joins the network. However, for some scenarios (e.g. satellite communications), characterizing the channel experienced by the received signal might be effective in detecting impersonation. In this work, we propose FadePrint, a solution capable of detecting satellite spoofing attacks by fingerprinting the noise-fading process associated with the satellite communication channel. The fading characteristics of a satellite link differ significantly from terrestrial links (e.g., indoor), making it possible to distinguish between the two. Unlike other systems, FadePrint does not require retraining when new transducers are added to the network. We tested FadePrint with real satellite and indoor radio measurements and proved that FadePrint can effectively discriminate between a satellite transmitter and a fake indoor one, with an accuracy higher than 0.99 for all the considered configurations.