To address the lack of systematic optimization strategies for trap relay deployment in the Tor network,this paper proposes a deployment framework utilizing multi-objective optimization and randomized obfuscation to enhance both de-anonymization efficiency and node stealthiness.Specifically,the paper,first constructs a local connectivity graph using the Tor control protocol, and integrates multiple centrality metrics,including degree,closeness,and Katz centrality.It then quantifies node monitoring visibility by applying min-max normalization and temporal weighting.Subsequently,the trap relay deployment problem is formulated as a multi-objective optimization model aimed at maximizing surveillance coverage,minimizing operational costs, and reducing detection risks.The NSGA-Ⅱ algorithm is then employed to generate Pareto-optimal solutions.Concurrently,a Poisson-distribution-based randomized obfuscation mechanism is introduced to dynamically adjust trap relay activation rates,thereby making their behavioral characteristics resemble those of legitimate relay nodes.Simulation experiments based on real Tor network topology and traffic data demonstrate that,under identical injection scales,the proposed approach significantly reduces the proportion of trap relays flagged as suspicious by SybilHunter while maintaining high monitoring visibility and traffic coverage.Compared with baseline strategies,this framework achieves an effective balance between stealthiness and resource utilization,demonstrating strong robustness and scalability.The results indicate that combining multi-objective optimization with dynamic obfuscation mechanisms can significantly enhance the survivability and de-anonymization potential of trap relays,providing a quantifiable deployment framework for anonymity system security evaluation and anonymity research.
DingledineR, MathewsonN, SyversonP.Tor: The second-generation onion router [C]//Proceedings of the 13th USENIX Security Symposium, 2004: 303-320.
[2]
SyversonP F, ReedM G, GoldschlagD M.Onion routing access configurations [C]//Proceedings of the DARPA Information Survivability Conference and Exposition(DISCEX'00), 2000: 34-40.
JardineE, LindnerA M, OwensonG.The potential harms of the Tor anonymity network cluster disproportionately in free countries [J].P Natl Acad Sci USA, 2020, 117(50): 31716-31721.
[7]
FeigenbaumJ, JohnsonA, SyversonP.A model of onion routing with provable anonymity [M]//Financial Cryptography and Data Security.Berlin,Heidelberg: Springer, 2007: 57-71.
[8]
DahalS, LeeJ, KangJ, et al.Analysis on end-to-end node selection probability in Tor network [C]//2015 International Conference on Information Networking (ICOIN).Cambodia: IEEE, 2015: 46-50.
[9]
ProjectTor.Entry guards [EB/OL].[2025-10-08].
[10]
LiQ, LiuP, QinZ.A stealthy attack against Tor guard selection [J].Int J Secur Appl, 2015, 9(11): 391-402.
[11]
TanQ, WangX, ShiW,et al.An anonymity vulnerability in Tor [J].IEEE ACM T Network, 2022, 30(6): 2574-2587.
[12]
ProjectTor.Relay operators [EB/OL].[2025-10-08].
[13]
LiangX, XiaoY.Game theory for network security [J].IEEE Commun Surv Tutorials, 2013,15(1): 472-486.
[14]
MurdochS J, DanezisG.Low-cost traffic analysis of Tor [C]//2005 IEEE Symposium on Security and Privacy (SP'05).Oakland, CA, USA: IEEE, 2005: 183-195.
[15]
JansenR, JuarezM, GalvezR, et al.Inside job: Applying traffic analysis to measure Tor from within [C]// Proceedings 2018 Network and Distributed System Security Symposium, 2018: 1-15.
[16]
MengR, YeY, XieN G.Multi-objective optimization design methods based on game theory [C]//Proceedings of the World Congress on Intelligent Control and Automation, 2010: 2220-2227.
[17]
AminuddinM A I M, FitriZ, KaurM, et al.A survey on Tor encrypted traffic monitoring [J].Int J Adv Comput Sc, 2018, 9(8): 113-120.
[18]
BasyoniL, FetaisN, ErbadA,et al.Traffic analysis attacks on Tor: A survey [C]//2020 IEEE International Conference on Informatics,IoT,and Enabling Technologies (ICIoT), 2020: 183-188.
[19]
KarunanayakeI, AhmedN, MalaneyR,et al.De-anonymisation attacks on Tor: A survey [J].IEEE Commun Surv Tut, 2021, 23(4): 2324-2350.
BauerK, McCoyD, GrunwaldD,et al.Low-resource routing attacks against Tor [C]//Proceedings of the 2007 ACM Workshop on Privacy in Electronic Society, 2007: 11-20.
DebK.Multi-objective optimisation using evolutionary algorithms:An introduction [M]//Multi-objective Evolutionary Optimisation for Product Design and Manufacturing.London: Springer London, 2011: 3-34.
[26]
DebK, PratapA, AgarwalS,et al.A fast and elitist multiobjective genetic algorithm: NSGA-Ⅱ[J].IEEE Trans Evol Computat, 2002, 6(2): 182-197.
[27]
JohnsonA, WacekC, JansenR, et al.Users get routed: Traffic correlation on Tor by realistic adversaries [C]//Proceedings of the 2013 ACM SIGSAC Conference on Computer Communications Security-CCS '13, 2013: 337-348.
[28]
JohnsonA, WacekC, JansenR, et al.TorPS: The Tor path simulator [EB/OL].[2025-10-08].
[29]
ProjectTor.Tor metrics [EB/OL].[2025-10-08].
[30]
WinterP, EnsafiR, LoesingK, et al.Identifying and characterizing Sybils in the Tor network [C]//25th USENIX Security Symposium (USENIX Security 16), 2016: 1169-1185.
[31]
WinterP, EnsafiR, LoesingK, et al.SybilHunter [EB/OL].[2025-10-08].