Back to search

Cornell University

Application Awareness in Censorship Circumvention Systems

Abstract

dc:description.abstract

Internet censorship circumvention systems rely on cover applications to conceal the existence of their communication: covert data is tunneled through the encrypted network channel of the cover applications to provide network-level unobservability. However, recently proposed systems are not aware of the cover application semantics, making them vulnerable to adversaries that can observe the application-level behavior of cover applications. In this thesis, I present my Ph.D. research on new classes of attacks on censorship circumvention systems that exploit the lack of application awareness in existing designs, as well as the mitigation of these attacks. We first demonstrate a new type of active attack we call "differential degradation." These attacks exploit the discrepancies between the network requirements of the cover application and those of the circumvention system, allowing adversaries to detect and block state-of-the-art systems that resist network traffic analysis-based attacks. Differential degradation doesn't require complex multi-flow measurement or traffic classification, making it feasible for realistic censors at low cost. I'll also discuss the root causes of these vulnerabilities and the trade-offs faced by designers of circumvention systems. Then, we show how a network-based adversary is able to observe a cover application’s events through encrypted traffic flows, and thus can detect the violation of application-specific invariants caused by tunneling data through the cover application using popular content substitution designs. To mitigate the problem, we propose non-disruptive content substitution that does not introduce inconsistencies into the application state. Using this idea, we design and implement Telepath, a Minecraft-based covert communication system that is immune to our attacks while providing resistance to traditional traffic analysis attacks.

Degree

thesis:*
Name thesis:degree_name
Ph. D., Computer Science
Level thesis:degree_level
Doctor of Philosophy
Discipline thesis:degree_discipline
Computer Science
Grantor
Cornell University
Year dc:date.issued
2024

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Sun, Zhen
Chair dc:contributor.chair
  • Shmatikov, Vitaly
Committee members dc:contributor.committeemember
  • Ristenpart, Thomas
  • Zabih, Ramin

Rights

Language dc:language.iso
en

Identifiers

dc:identifier.*
Dc Identifier Bibid
16611786
Dc Identifier Other
ProQuest Submission ID: 14447
ProQuest Publication ID: 31482931

Chain of custody

source
Harvested from
Cornell University
Base URL
ecommons.cornell.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Sun, Zhen. Application Awareness in Censorship Circumvention Systems. Doctor of Philosophy thesis, Cornell University, 2024. https://hdl.handle.net/1813/116591