Secure Distance Bounding Verification using Physical-Channel Properties
This addresses the problem of secure distance verification in wireless systems, offering an alternative to clock-based methods, but it is incremental as it builds on existing fraud attack models and introduces a conditional solution for terrorist fraud.
The paper tackles secure distance bounding verification by using physical channel properties instead of time measurement, showing that efficient protocols can achieve security against distance and mafia fraud attacks even against unbounded adversaries, but terrorist fraud security is impossible without time measurement unless the adversary's communication is limited to the bounded retrieval model.
We consider the problem of distance bounding verification (DBV), where a proving party claims a distance and a verifying party ensures that the prover is within the claimed distance. Current approaches to "secure" distance estimation use signal's time of flight, which requires the verifier to have an accurate clock. We study secure DBV using physical channel properties as an alternative to time measurement. We consider a signal propagation environment that attenuates signal as a function of distance, and then corrupts it by an additive noise. We consider three attacking scenarios against DBV, namely distance fraud (DFA), mafia fraud (MFA) and terrorist fraud (TFA) attacks. We show it is possible to construct efficient DBV protocols with DFA and MFA security, even against an unbounded adversary; on the other hand, it is impossible to design TFA-secure protocols without time measurement, even with a computationally-bounded adversary. We however provide a TFA-secure construction under the condition that the adversary's communication capability is limited to the bounded retrieval model (BRM). We use numerical analysis to examine the communication complexity of the introduced DBV protocols. We discuss our results and give directions for future research.