Introduction
Contactless technology, such as RFID and NFC, relies on an implicit security assumption: if a reader can communicate with a token (e.g., a credit card), the two must be in close physical proximity, typically less than 10 cm. A relay attack, also known as a "wormhole attack," breaks this assumption by extending the communication range using intermediate proxies. Historically, these attacks required specialized hardware. This research demonstrates that standard, off-the-shelf NFC smartphones are now capable of executing these attacks stealthily and effectively.
The Relay Mechanism
The attack involves two proxy devices that tunnel communication over a long-distance channel, such as Bluetooth or Wi-Fi. A "proxy-reader" is placed near the victim's legitimate token, while a "proxy-token" is placed near the target reader.
The reader believes it is interacting directly with the victim's token because the proxies transparently forward the Application Protocol Data Units (APDUs). This bypasses application-layer security, including encryption and mutual authentication, because the attacker does not need to decrypt the data—they simply relay the bits.
Implementation: The "Soft-SE" Breakthrough
The researchers implemented this system using a Nokia 6131 (proxy-reader) and a BlackBerry 9900 (proxy-token). The critical innovation was the use of the BlackBerry's NFC emulation API to create a "soft-SE" (Software Secure Element). This allowed the phone to mimic a passive contactless token entirely in software, eliminating the need for custom hardware or programmable smart cards.
The relay channel utilized Bluetooth (L2CAP protocol) for its low setup complexity. The system was designed to be application-independent, acting as a "bit-pipe" capable of relaying any ISO 14443-4 compliant transaction, from credit card payments to e-passport scans.
Experimental Results and the Latency Myth
A common misconception in NFC security is that strict timing requirements would prevent relay attacks over relatively slow channels like Bluetooth. However, experimental data reveals that real-world readers are surprisingly "patient."
While a direct card-to-reader transaction might take between 4ms and 17ms per command, the relay setup added approximately 50ms of round-trip latency. Despite this overhead, the researchers successfully completed transactions because commercial Point-of-Sale (POS) terminals often wait up to 35 seconds for a response, and e-passport readers wait over 5 seconds.
The attack proved reliable at distances of up to 15 meters through walls and 35 meters in open spaces. The use of standard smartphones also provides a high degree of stealth, as an attacker appears to be simply using their phone near a payment terminal.
Countermeasures and Future Work
To defend against such attacks, the authors suggest several strategies:
- Distance Bounding: Implementing protocols that measure the Round Trip Time (RTT) at the physical layer to ensure the token is within a few centimeters. This is the most robust defense but requires hardware updates.
- Location-Based Verification: Using the phone's GPS or network location as a second factor. If the proxy-reader and proxy-token are in different locations, the transaction can be flagged.
- Ambient Sensing: Comparing the ambient environment (e.g., audio, light, or Wi-Fi signals) at both ends of the relay to detect discrepancies.
While proprietary protocols like Mifare Classic remain difficult to relay due to non-standard framing, the move toward standardized ISO 14443-4 protocols makes the "soft-SE" relay attack a significant and practical threat to modern contactless infrastructure.
Summary updated