Design and implementation of a terrorist fraud resilient distance bounding system

Abstract

Given the requirements of fast processing and the complexity of RF ranging systems, distance bounding protocols have been challenging to implement so far; only few designs have been proposed and implemented. Currently, the most efficient implementation of distance bounding protocols uses analog processing and enables the prover to receive a message, process it and transmit the reply within 1 ns, two orders of magnitude faster than the most efficient digital implementation. However, even if implementing distance bounding using analog processing clearly provides tighter security guarantees than digital implementations, existing analog implementations do not support resilience against Terrorist Fraud attacks; they protect only against Distance Fraud and Mafia Fraud attacks. We address this problem and propose a new, hybrid digital-analog design that enables the implementation of Terrorist Fraud resilient distance bounding protocols. We introduce a novel attack, which we refer to as the "double read-out" attack and show that our proposed system is also secure against this attack. Our system consists of a prototype prover that provides strong security guarantees: if a dishonest prover performs the Terrorist Fraud attack, it can cheat on its distance bound to the verifier only up to 4.5 m and if it performs Distance Fraud or Mafia Fraud attacks up to 0.41 m. Finally, we show that our system can be used to implement existing (Terrorist Fraud resilient) distance bounding protocols (e.g., the Swiss Knife and Hancke-Kuhn protocol) without requiring protocol modifications. © 2012 Springer-Verlag.