I am broadly interested in systems and network security and Privacy Enhancing Technologies (PETs). My current interests are in Anonymous Communication Networks like Tor and Mix networks, with an emphasis on effective, efficient, and robust deployments. I am also active in the area of Data-Driven Privacy Systems Engineering where the use of telemetry and contextual data provide inputs into an engineering approach to system and network design and optimisation. More recently, I am conducting transitional research to Embedding Privacy into the Infrastructure based on my prior work on Tor, Mix networks, and privacy-preserving data acquisition and analytics.
Tor is a most popular anonymous communication network in the world today with over 2 million daily users. However, its security and performance characteristics are an active research area. Our research proposals have been adopted by Tor leading to improved client security and have proposed award-winning schemes to make Tor faster. We have also continue to investigate censorship resistance thorough Tor and beyond building on our taxonomy and game-theoretic frameworks.
While Tor is popular, it has limitation such as being insecure in the face of a global adversary. Our new mix network design, Loopix, is faster than prior designs and has been adopted by the secure messaging community.
Sample relevant publications:
AM Piotrowska, J Hayes, T Elahi, S Meiser, and G Danezis. "The Loopix Anonymity System." USENIX Security Symposium, 2017.
S Khattak, T Elahi, L Simon, CM Swanson, SJ Murdoch, and I Goldberg. "SOK: Making sense of censorship resistance systems." Privacy Enhancing Technologies (PETS), 2016.
T Elahi, JA Doucette, H Hosseini, SJ Murdoch, and I Goldberg. "A framework for the game-theoretic analysis of censorship resistance." Proceedings on Privacy Enhancing Technologies (PETS), 2016.
Systems engineering requires modelling but the data required can often be sensitive. Privacy-preserving modelling and Private Data Analysis provide tools to both learn how systems and users behave without risks to privacy. We are investigating applications of machine learning techniques to PETS design and evaluation with current interests in feedback control and optimisation.
Sample relevant publications:
L Watson, A Mediratta, T Elahi, R Sarkar. "Privacy Preserving Detection of Path Bias Attacks in Tor". To appear in Proceedings on Privacy Enhancing Technologies 2020, no. 4 (2020).
R Jansen, M Juarez, R Galvez, T Elahi, C Diaz. Inside Job: Applying Traffic Analysis to Measure Tor from Within. Proceedings of the Network and Distributed System Security (NDSS), 2018.
T Elahi, G Danezis, and I Goldberg. "Privex: Private collection of traffic statistics for anonymous communication networks." Proceedings of the ACM Computer and Communications Security (CCS), 2014.
We are adapting Tor and Mixnets to function as a Network Privacy Layer to bring privacy and anonymity to a broader range of users and applications and to make subverting privacy more difficult. Our mixnet design, Loopix, has been incorporated to anonymous blockchains and we are currently investigating privacy mechanisms to leverage in the IoT setting.
Sample relevant publications:
AM Piotrowska, J Hayes, T Elahi, S Meiser, and G Danezis. "The Loopix Anonymity System." USENIX Security Symposium, 2017.
M Alsabah, K Bauer, T Elahi, Ian Goldberg. "The Path Less Travelled: Overcoming Tor's Bottlenecks with Traffic Splitting." Privacy Enhancing Technologies Symposium, 2013. [pdf]
Hazen Li (Visiting PhD student, 2023-2024)
Lauren Watson (PhD 2024, second supervisor)
Neenu Garg (Postdoc, 2023-present)
Kaushik Halder (Postdoc, 2021-2022)
Ibrahim Ahmed (PhD 2022, second supervisor)
Florentin Rochet (Postdoc, 2021-2022)
Xinshu Ma (PhD candidate, 2021-2023)
Jenjira Jaimunk (Postdoc, May-September 2021, co-superviser)
Vangelis Dimitriou, "Measuring Performance Effect of Tor Bandwidth Publishing Intervals" (19-20)
Aristide Fumagalli , "TorSim: Customizable Emulation Environment for Realistic Tor Experimentation" (19-20)
Ventsislav Kostadinov, "Mix Network Test-bed Setup for Experimentation" (19-20)
Gabriel Mawhinney, "Website Fingerprinting on Tor Pluggable Transports" (19-20)
Alexander Pietz, "Next Generation Onionoo Service: Using Time-Series Forecasting to Determine Guard Relay Eligibility" (18-19)