An Adversary Model for Simulation-Based Anonymity Proof

The use of a formal method is a promising approach to developing reliable computer programs. This paper presents a formal method for anonymity, which is an important security property of communication protocols with regard to a user's identity. When verifying the anonymity of security protocols, we need to consider the presence of adversaries. To formalize stronger adversaries, we introduce an adversary model for simulation-based anonymity proof. This paper also demonstrates the formal verification of a communication protocol. We employ Crowds, which is an implementation of an anonymous router, and verify its anonymity. After describing Crowds in a formal specification language, we prove its anonymity with a theorem prover.