TY - GEN
T1 - HiddenTor
T2 - 44th IEEE International Conference on Distributed Computing Systems, ICDCS 2024
AU - Zang, Yichen
AU - Cai, Chengjun
AU - Dong, Wentao
AU - Xu, Lei
AU - Wang, Cong
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - Tor bridges are crucial, unlisted relays designed to enhance system accessibility and circumvent censorship in the Tor network. Currently, Tor BridgeDB will randomly distribute 1-3 bridge relays to the user per request. Yet, those randomly selected bridges may not meet users' specific needs, e.g., adequate bandwidth for large-file sharing in a certain region. Also, a user's usage metadata (e.g., bridge choices) collected by Tor BridgeDB would inevitably reveal sensitive information about the user, discouraging the use of this censorship-circumvention service. In light of them, we introduce HiddenTor, a user-centric and privacy-focused bridge distribution system that allows Tor users to retrieve bridges privately and precisely (i.e., based on a range of specific criteria). At its core, HiddenTor designs a condition-based private information retrieval (PIR) protocol by building atop a suite of lightweight cryptographic primitives (i.e., function secret sharing). Besides, HiddenTor also crafts several optimization designs to balance the trade-offs between query efficiency and service reliability. The extensive experimental results have confirmed the feasibility and practicality of HiddenTor. For example, our prototype can efficiently handle private queries over 3000 bridges in approximately 2 seconds, which can further be reduced to 0.21 seconds using parallel computing techniques.
AB - Tor bridges are crucial, unlisted relays designed to enhance system accessibility and circumvent censorship in the Tor network. Currently, Tor BridgeDB will randomly distribute 1-3 bridge relays to the user per request. Yet, those randomly selected bridges may not meet users' specific needs, e.g., adequate bandwidth for large-file sharing in a certain region. Also, a user's usage metadata (e.g., bridge choices) collected by Tor BridgeDB would inevitably reveal sensitive information about the user, discouraging the use of this censorship-circumvention service. In light of them, we introduce HiddenTor, a user-centric and privacy-focused bridge distribution system that allows Tor users to retrieve bridges privately and precisely (i.e., based on a range of specific criteria). At its core, HiddenTor designs a condition-based private information retrieval (PIR) protocol by building atop a suite of lightweight cryptographic primitives (i.e., function secret sharing). Besides, HiddenTor also crafts several optimization designs to balance the trade-offs between query efficiency and service reliability. The extensive experimental results have confirmed the feasibility and practicality of HiddenTor. For example, our prototype can efficiently handle private queries over 3000 bridges in approximately 2 seconds, which can further be reduced to 0.21 seconds using parallel computing techniques.
KW - Function Secret Sharing
KW - Private Query
KW - Tor
UR - https://www.scopus.com/pages/publications/85203140930
U2 - 10.1109/ICDCS60910.2024.00119
DO - 10.1109/ICDCS60910.2024.00119
M3 - 会议稿件
AN - SCOPUS:85203140930
T3 - Proceedings - International Conference on Distributed Computing Systems
SP - 1260
EP - 1271
BT - Proceedings - 2024 IEEE 44th International Conference on Distributed Computing Systems, ICDCS 2024
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 23 July 2024 through 26 July 2024
ER -