Deep-Learning-Aided RF Fingerprinting for NFC Relay Attack Detection

Near field communication (NFC) has been a widely used radiofrequency identification (RFID) technology, credited to its convenience and security features. However, the transmitted signals can be easily eavesdropped or relayed in an open wireless channel. One of challenges is relay attack, where attacker simply relays signal bypasses encryption other means application layer. Prior works on attack countermeasures have focused distance-bounding protocols ambient-based solutions. This paper focuses ISO/IEC 14443-A proposes NFC detection method based RF fingerprinting physical To this end, we first designed implemented two realizations attacks, wired relays, built SDR-based testbed. We collected normal four tags, answer request type A (ATQA) segments were selected for fingerprinting. The created dataset comprised 66,366 samples, with tags’ signals. was then fed into deep CNN training. Finally, our experiment results showed that effectively distinguished high accuracy 99%, confirming promising countermeasure attacks.