Traffic Analysis Attacks in Anonymity Networks

Abstract

With more than 1.7 million daily users, Tor is a large-scale anonymity network that helps people to protect their identities in the Internet. Tor provides low-latency transmissions that can serve a wide range of applications including web browsing, which renders it an easily accessible tool for a large user base. Unfortunately, its wide adoption makes Tor a valuable target for de-anonymization attacks. Recent work proved that powerful traffic analysis attacks exist which enable an adversary to relate traffic streams in the network and identify users and accessed contents. One open research question in the field of anonymity networks therefore addresses efficient countermeasures to the class of traffic analysis attacks. Defensive techniques must improve the security features of existing networks while still providing an acceptable performance that can maintain the wide acceptance of a system. The proposed work presents an analysis of mixing strategies as a countermeasure to traffic analysis attacks in Tor. First simulation results indicate the security gains and performance impairments of three main mixing strategies.