When proximity is not presence
Summary: A radio signal may suggest that a device is nearby. That alone does not establish that two devices actually share the same physical situation.
Many digital systems treat proximity as a security property. A contactless key unlocks a vehicle. A badge opens a door. A phone authorizes an interaction because it appears to be within short range.
This assumption is useful. It is not equivalent to proof of presence.
Proximity, distance and co-presence are different claims
Proximity describes an estimated spatial relationship between objects. Depending on the technology, it may be inferred from received signal strength, propagation time, transmitter visibility or other channel properties.
Co-presence, in a security context, asks a stronger question: are the devices actually participating in the same physical situation, at the same place and time?
The distinction matters because a measurement is not the property it is intended to represent.
A high RSSI, for example, means that a receiver observes a relatively strong signal. Received power also depends on transmit power, antennas, orientation, obstacles, reflections and multipath propagation. RSSI can therefore be useful as a proximity indicator, but by itself it is not cryptographic proof of distance or co-presence.
Proximity is a measurement. Co-presence is a security claim.
The relay-attack problem
Relay attacks expose this distinction directly.
They do not necessarily require an attacker to break encryption or learn the secrets held by legitimate devices. Instead, communication can be transported between legitimate but distant endpoints so that each endpoint interacts as though the other were within range.
Francillon, Danev and Čapkun demonstrated this on Passive Keyless Entry and Start systems. Their experiments across ten car models from eight manufacturers showed that relaying communications could enable entry and starting while the genuine key remained distant. The authors also noted that the relay was independent of the presence of strong authentication and encryption.
The broader lesson is important: a legitimate identity can participate in an interaction whose apparent physical topology is false.
Why short range does not mean non-relayable
Bluetooth, NFC and other wireless technologies have an expected physical range. Range naturally constrains ordinary communication, but it should not be confused with absolute proof of presence.
NIST SP 800-63B explicitly notes that wireless authenticator connections can be vulnerable to eavesdropping, injection and relay attacks. Nominal communication range is therefore part of the threat model, not a complete solution to it.
Research on Zero-Interaction Authentication reaches a related conclusion. Truong et al. describe systems relying on the detection of a personal device over a short-range wireless channel as vulnerable to practical relay attacks, and investigate contextual information as a mechanism for co-presence detection.
Measuring distance is still a different question
More advanced techniques can estimate physical distance, including methods based on propagation time. Such mechanisms can provide substantially stronger properties than received signal strength alone.
The concepts should nevertheless remain distinct.
Distance estimation primarily asks:
How far away does the other device appear to be?
Co-presence verification asks:
Do we have sufficient evidence to conclude that these devices actually participate in the same physical situation?
The questions are related. They are not identical.
The environment as security information
Another body of research considers shared physical context.
The intuition is that genuinely co-present devices may observe common properties of their environment, while a remote device observes a different context.
Conti et al. use the term Contextual Co-presence (COCO) for techniques that use contextual information to improve proximity verification. Their survey also stresses that no ideal solution exists: security, efficiency, deployment cost and usability involve trade-offs.
This prevents a simplistic conclusion such as "context always proves presence." The strength of any evidence depends on the observed phenomenon, the adversary model, how reproducible the context is, and the protocol that uses it.
The question Keypulse investigates
This distinction defines the research problem explored by Keypulse.
The objective is not merely to determine whether a device appears nearby. It is to investigate how a digital interaction can obtain evidence about a property of the physical world: co-presence.
That changes the security question.
Instead of asking only:
"Who is this device?"
or:
"How far away does it appear to be?"
we can also ask:
"What establishes that we are actually here, together, now?"
Answering that question rigorously requires separating radio measurements, distance, context and evidence of co-presence. That distinction is the foundation for the next articles in this series.
References
- NIST, SP 800-63B --- Digital Identity Guidelines: Authentication and Authenticator Management, wireless connections. https://pages.nist.gov/800-63-4/sp800-63b/authenticators/
- Aurélien Francillon, Boris Danev, Srdjan Čapkun, Relay Attacks on Passive Keyless Entry and Start Systems in Modern Cars, NDSS, 2011. https://eprint.iacr.org/2010/332
- Hien Thi Thu Truong et al., Comparing and fusing different sensor modalities for relay attack resistance in Zero-Interaction Authentication, IEEE PerCom, 2014, DOI: 10.1109/PerCom.2014.6813957. https://research.aalto.fi/en/publications/comparing-and-fusing-different-sensor-modalities-for-relay-attack/
- Mauro Conti et al., Context-based Co-presence Detection Techniques: A Survey, Computers & Security, 2019, DOI: 10.1016/j.cose.2019.101652. https://www.sciencedirect.com/science/article/pii/S0167404819301968