STARS: A Statistical Traffic Pattern Discovery System for Anonymous MANET Communications

Anonymous MANET routing relies on techniques such as re-encryption on each hop to hide end-to-end communication relations. However, passive signal detectors and traffic analyzers can still retrieve sensitive information from PHY and MAC layers to derive end-to-end communication relations through statistical traffic analysis. In this paper, the authors propose a Statistical Traffic pAttern discoveRy System (STARS) based on eigen analysis which can greatly improve the accuracy to derive traffic patterns in MANETs. STARS intends to find out the sources and destinations of captured packets and to discover the end-to-end communication relations. The proposed approach is purely passive.