Securing Semantic Communications with Physical-layer Semantic Encryption and Obfuscation. (arXiv:2304.10147v1 [cs.CR])

Deep learning based semantic communication(DLSC) systems have shown great
potential of making wireless networks significantly more efficient by only
transmitting the semantics of the data. However, the open nature of wireless
channel and fragileness of neural models cause DLSC systems extremely
vulnerable to various attacks. Traditional wireless physical layer key (PLK),
which relies on reciprocal channel and randomness characteristics between two
legitimate users, holds the promise of securing DLSC. The main challenge lies
in generating secret keys in the static …

attacks challenge channel communication communications data deep learning encryption environment great key keys lies low main making nature networks obfuscation physical randomness rate secret secret keys systems vulnerable wireless