flow-correlation   Flow-correlation traffic analysis

A three-stage detection pipeline exploiting the "dual-role" behavioral fingerprint of single-IP circumvention relays achieved 23.2% recall (96/414 ground-truth relays) with a 0.18% false-positive rate against 97,651 benign TLS servers, for an overall accuracy of 99.5%. The ground-truth set covered OpenVPN, WireGuard, and SOCKS relays identified in a 17 TB single-day backbone trace (WIDE Project, April 9, 2025).

2026-almutairi-server §3.4, Table 1 detection

In the 30 November 2025 Onionoo snapshot, approximately 75% of both guard and middle relay selection-probability mass lies within Fourteen-Eyes jurisdictions, with Germany, the United States, and the Netherlands hosting the largest shares. Because path selection is bandwidth-weighted, a state-level adversary from one of these jurisdictions is statistically likely to have ISP-level visibility at most relays a target introduction circuit passes through.

2026-constantinides-cost-stability-deanonymizing Appendix D, Fig. 7 evaluation

An adaptive intersection attack exploiting the 18–24h fixed lifetime of Tor onion-service introduction circuits reconstructed the complete four-hop introduction circuit in all nine live-network experiments. The median reconstruction required 260 iterations with an estimated median wall-clock time of 2.24h (12% of the 18h circuit lifetime) when per-relay visibility cost v=0; the slowest run completed in 6.63h (37% of 18h).

2026-constantinides-cost-stability-deanonymizing §4.2, Table 1 detection

A persistent 'tail' of recurring candidates — not the early phase — dominates convergence cost and introduces high variance: three Vanguard stages at identical consensus weight 1200 converged at iterations 12, 142, and 251. In 13 of 36 stages, the intersection held ≥3 candidates at the penultimate iteration and collapsed to a singleton in one step, suggesting the tail is caused by persistent third-party guard/vanguard traffic rather than random noise.

2026-constantinides-cost-stability-deanonymizing §4.2, Table 2 detection

Across all 36 stages in the nine end-to-end experiments, the median iteration at which the running intersection shrank to ≤10 candidates was 3, and 64% of stages reached this threshold within 5 iterations. Per-stage candidate sets shrank by 72–87% after the first intersection and by 85–96% by the third iteration, driven by the short (~0.5–1.8s) observation windows that prevent unrelated circuits from injecting persistent false candidates.

2026-constantinides-cost-stability-deanonymizing §4.2 detection

Periodically rebuilding only the internal service-side path of the introduction circuit (entry guard → vanguard → middle relay) approximately every 10 minutes — while keeping the publicly advertised Introduction Point fixed — limits evidence accumulation to one rebuild interval without requiring a descriptor update. Under this mitigation the effective circuit lifetime visible to an intersection attacker drops from 18–24h to ~10 min, matching ordinary Tor circuit lifetimes.

2026-constantinides-cost-stability-deanonymizing §5 defense

Monte Carlo simulations across four EoL exclusion rounds (2021–2023) show that natural relay churn is the dominant driver of Tor client anonymity loss, with EoL exclusions giving AS and IXP adversaries an advantage equivalent to only ~4 days (AS adversary) or ~9 days (relay adversary) of the network running without exclusion. In the 2021 exclusion event, up to ~1% of additional clients were compromised; by 2023 this figure dropped to ~0.30%, attributed to improved operator update discipline and Tor Project communications.

2026-dejaeghere-sociotechnical-aspects-tor §5.1–5.3, Figure 5–7 evaluation

NATA (Non-invasive Active Traffic-correlation Analysis) injects low-frequency bandwidth waveforms (sinusoidal, square-wave, triangular) into Tor TCP connections at an upstream gateway without endpoint compromise, payload decryption, or Tor-browser modification. BM-Net, a selective state-space classifier trained on the exit-side observations, achieves a 99.65% binary detection F1 score distinguishing watermarked from natural traffic on a 20,000-trace real-world dataset.

2026-fan-activeflowmark-assessing-tor §IV, §VI.D, Table III detection

BM-Net achieves a 99.65% binary detection F1 score for distinguishing bandwidth-watermarked Tor flows from natural traffic, outperforming all evaluated baselines (next best: TikTok at 75.96% F1). The accuracy gap stems from active perturbation imposing a deterministic low-frequency throughput constraint rather than relying on subtle natural metadata, making the detection task fundamentally easier than passive website fingerprinting.

2026-fan-activeflowmark-assessing-tor §VI-D, Table III–IV detection

BM-Net achieves a 99.65% binary detection F1 score distinguishing watermarked from natural Tor flows, and a 97.5% macro-F1 score for fine-grained modulation classification across sinusoidal, square-wave, and triangular patterns. The fine-grained test set contains 201 held-out samples collected from ten clients across five geographic regions (Europe, North America, Australia, Southeast Asia, East Asia), with training traces including traffic collected under WTF-PAD and Walkie-Talkie defenses.

2026-fan-activeflowmark-assessing-tor §VI-D, Table III, Table V evaluation

Active bandwidth perturbation has an inherent detectability–stability trade-off: overly aggressive low-rate phases cause Tor SENDME-based flow control stalls, retransmissions, timeouts, or circuit replacement before sufficient correlation evidence is collected. The paper selects a 30-second modulation period and an empirically determined minimum shaping rate; the usable shaping range varies with relay load, path length, TCP congestion control behavior, and Tor multiplexing.

2026-fan-activeflowmark-assessing-tor §VII-A detection

Tornettools-based simulations with a 1%-scaled Tor network show that with 1 adversary-controlled exit relay (148 Mbps), the exit-observation probability pexit ≈ 2.13%; with 5 adversary-controlled relays, pexit exceeds 10%. Tor's bandwidth-weighted path selection means high-bandwidth malicious exits attract a disproportionate share of circuits, and repeated observations over multiple circuits compound correlation risk multiplicatively even at modest relay counts. The simulation scaled historical Tor consensus data to 1% of the real network (80 relays), and repeated observations across T independent windows compound as 1 - (1 - Pcorr)^T.

2026-fan-activeflowmark-assessing-tor §VI-A, Table I evaluation

BM-Net achieves a 97.5% macro-F1 score for fine-grained classification of three modulation geometries (sinusoidal, square-wave, triangular) from noisy exit-side Tor observations using only 201 labeled test samples collected across cross-continental Tor paths. Residual errors concentrate between natural traffic and square-wave modulation, as abrupt low-rate transitions are partially smoothed by Tor multiplexing and network jitter. Per-class F1 is 95.7% for square-wave, the hardest class, versus 99.0% each for sinusoidal and triangular.

2026-fan-activeflowmark-assessing-tor §VI-D, Table V detection

NATA requires no endpoint compromise, no Tor-browser modification, and no payload decryption; it operates solely from (1) an upstream gateway controlling Tor TCP connections via standard Linux tc/wondershaper rate-limiting and (2) one or more adversary-controlled exit relays passively recording packet traces. The shaper identifies Tor connections using flow-level metadata (client IP, relay IP, port, transport protocol), meaning the adversary needs only ISP or AS-level vantage, not host-level access. The authors name this Shaper-Sniffer architecture NATA, for Non-invasive Active Traffic-correlation Analysis.

2026-fan-activeflowmark-assessing-tor §III-A, §IV-A detection

Padding-based client-side defenses including WTF-PAD and Walkie-Talkie are insufficient against active bandwidth perturbation: they reshape packet timing and burst structure but cannot remove the upstream rate limit imposed by the gateway shaper. BM-Net trained on a defense-aware dataset containing both undefended and WTF-PAD/Walkie-Talkie traces still achieves 99.65% F1, and the paper explicitly notes that 'client-side padding and burst reshaping may alter the logical traffic pattern, but they do not directly remove the rate limit imposed by the upstream bottleneck.'

2026-fan-activeflowmark-assessing-tor §VI.E.2, §VII detection

Multi-censor simulations show that single-censor-optimized distribution strategies perform suboptimally in realistic multi-region deployments. When two networks have different censor strategies (e.g., one optimal, one zig-zag), the distributor cannot detect that a proxy is blocked until all censors have blocked it; this leaves clients without reachable proxies despite the proxy appearing "available" from the distributor's view. The authors conclude that "single-censor evaluation does not accurately predict more realistic deployment performance." A zig-zag censor in one region with 0.25 weight caused 44.4% collateral damage while reducing proxy lifetime to a median of 4 steps.

2026-fares-game §6, Table 4 evaluation

The zig-zag traffic analysis attack (confirmed supported in Geedge TSG leak) rapidly enumerates all static proxy pools. With ζ_watch ∈ {4, 6} steps and a best-quality classifier (ρ_TP=0.99, ρ_FP=0.001), almost total proxy enumeration and user blockage occurs well before step 300. Even ζ_watch=2 leaves ~50% of users blocked. Collateral damage is high across all settings when ζ_watch ≥ 4: eventually ~50% of innocent servers are also blocked. However, Snowflake-style ephemeral proxies resist zig-zag effectively: reachability remains above 95% after 360 steps because churn prevents the censor from expanding its known proxy set beyond agents' direct assignments.

2026-fares-game §5.1, Fig 3, Fig 4 detection

Automated proxy engines (e.g., Xray-core running VLESS Reality in automated mode) generate deterministically rigid inter-arrival time distributions because they cannot synthesize the stochastic variance of human-driven IAT, even when volumetrically anchored to benign distributions ('Fat Middle' anchoring via AMOI). The AEGIS Thermodynamic Variance Detector identifies this rigidity via Shannon Entropy of hidden states across 1,000-packet causal windows, rendering volumetric anchoring mathematically distinguishable from genuine human traffic.

2026-ferrel-aegis-adversarial-entropy-guided §III-E, §V-B detection

At operationally realistic base rates—1 million connection pairs per hour with only 10 true stepping-stone chains—a detector with a 1% FPR generates approximately 10,000 false alarms per hour while correctly flagging all 10 intrusions, making classical statistical methods (which cannot reach FPR ≪ 10⁻²) operationally unusable; deep learning methods must target FPR ≤ 10⁻³ to be viable.

2026-mathews-tracing-chain-deep §I evaluation

ESPRESSO achieves only TPR 0.132 at FPR ≤ 10⁻³ in network-mode for DNS-tunneled traffic—near chance—compared to TPR 0.992 for SSH traffic at the same threshold. The paper attributes this to the polling-based communication mechanism of dnscat2, which disrupts the timing patterns that interval-based flow correlation relies on.

2026-mathews-tracing-chain-deep §V-B, Table III detection

ESPRESSO, a deep learning flow correlator combining a transformer backbone with time-aligned interval features and online triplet mining, achieves TPR >0.99 at FPR ≤ 10⁻³ for SSH, SOCAT, and ICMP stepping-stone traffic in network-mode detection, versus DCF's TPR of 0.320–0.956 across those same protocols at the same threshold. On the harder mixed-protocol dataset in network-mode, ESPRESSO achieves TPR 0.748 at FPR ≤ 10⁻³, more than double DCF's 0.334.

2026-mathews-tracing-chain-deep §V-B, Table III evaluation

Ablation experiments show that replacing ESPRESSO's transformer backbone with a CNN ('Modified DCF') while retaining time-aligned interval features achieves performance competitive with the full ESPRESSO model across most protocols (e.g., SOCAT network-mode pAUC 0.997 vs. 0.989 at FPR ≤ 10⁻³), demonstrating that the time-interval feature representation—not the transformer architecture—is the primary driver of correlation accuracy.

2026-mathews-tracing-chain-deep §V-B, Table III evaluation

A systematic robustness evaluation found that ESPRESSO is highly robust to packet padding alone but that even modest artificial timing jitter causes significant performance degradation, identifying timing-based perturbations as the primary vulnerability of correlation-based stepping-stone (and by extension, anonymity-network) detectors.

2026-mathews-tracing-chain-deep §I (contributions), Abstract detection

Fano's inequality establishes a theoretical lower bound on deanonymization error probability as a function of anonymity set size |Θ|, prior uncertainty H(X), and mutual information leakage I(X;Y). For a network of N sufficiently large nodes with uniform routing, Pe ≥ (log N − 1) / log(N−1), approaching 1 (perfect anonymity). The authors found that closed-form estimation of I(X;Y) from I2P traffic features was analytically intractable, requiring ML approximation — and that ML also failed in practice.

2026-rohrer-convolutional-neural-networks-deanonymisation-i2p §III-A evaluation

Applying Fano's inequality, the paper proves Pe ≥ (H(X)−1)/log|Θ|, showing that deanonymization error rate approaches 1 (perfect anonymity) when the anonymity set |Θ| is large and mutual information leakage I(X;Y) between observed traffic Y and target identity X is minimized. A uniform default tunnel length of 3 hops across all nodes, for example, contributes no differential leakage because p(y=3)=1, illustrating that standardized network parameters reduce identifiability.

2026-rohrer-convolutional-neural-networks-deanonymisation-i2p §III-A, Equation (2) detection

The Fragging attack exploits message fragmentation in Nym to break sender-recipient unlinkability without any external side information. In a Shadow network simulator testbed, benign fragment sets complete in an average of 308 ms, while sets with at least one dropped fragment show completion times of ~2000 ms due to Nym's retransmission timeout—a timing signal that remains distinguishable at per-hop mixing delays of up to 500 ms.

2026-schadt-breaking-partially-fixing §4.1, Figure 2 detection

Three conditions are jointly necessary and sufficient for a Fragging attack: (C1) the adversary can drop packets while knowing their sender—met by a malicious gateway/entry node in both Nym and Tor; (C2) the adversary can detect a missing fragment—met by Nym's explicit fragment-set identifiers and by Tor's encryption-state desynchronization on cell drop; and (C3) the adversary can learn the recipient from the surviving fragments—met in both systems because other fragments of the set arrive intact.

2026-schadt-breaking-partially-fixing §4.3 detection

Fragging-like attacks on Tor allow an adversary controlling the entry node to encode at least a 1-bit confirmation signal by dropping a single stream cell and observing circuit teardown (caused by encryption-state desynchronization between client and exit relay). If the adversary knows the expected cell count, they can select which cell to drop and thereby encode log(n) bits of information to the exit relay.

2026-schadt-breaking-partially-fixing §4.2 detection

Scylla, a modified Sphinx mix format, applies OAEP (an all-or-nothing transform) to the concatenated recipient address and message before splitting into fragments, so the final hop cannot recover the recipient or message without all fragments. Scylla adds only 39 bytes of header overhead over Sphinx and ~247 µs per fragment of computation time—negligible compared to Nym's default 15 ms per-hop mixing delay.

2026-schadt-breaking-partially-fixing §5, §7.1, §7.4, Figure 11 defense

Country-level analysis shows significant sensitivity of AS-level transit dependency to the distribution of relay capacity across ASes within each country, implying that a censor controlling a dominant national transit AS gains disproportionately high visibility into Tor traffic if domestic relay capacity is concentrated in that AS's customer cone.

2026-wang-who-carries-tor Abstract evaluation

The authors propose a diversity-aware relay placement heuristic derived from bandwidth-weighted AS Hegemony scores to guide future mitigation of transit concentration. By incorporating consensus bandwidth weighting into hegemony calculations, the methodology reflects actual Tor traffic patterns rather than relay counts alone.

2026-wang-who-carries-tor Abstract defense

SURF (AS1103) surged to over 50% exit hegemony in 2024, meaning a single AS gained visibility into the majority of bandwidth-weighted Tor exit traffic. This represents a dramatic concentration event in an otherwise persistent but distributed chokepoint landscape spanning 2015–2025.

2026-wang-who-carries-tor Abstract evaluation

A longitudinal study of Tor transit dependency from 2015 to 2025 reveals persistent concentration in a small number of transit ASes, indicating that AS-level chokepoints enabling flow-correlation attacks are a structural rather than transient property of the Tor network.

2026-wang-who-carries-tor Abstract evaluation

CenTor protects origin onion service operators from DoS and deanonymization by routing all client traffic through geographically distributed Bento replicas running inside SGX-based Trusted Execution Environments (TEEs). The original operator can go fully offline after deploying static content; replicas enforce confidentiality and integrity of hosted content with ephemeral per-enclave encryption keys, preventing malicious Bento node operators from inspecting or modifying content even if they control the underlying hardware.

2025-arora-improving-performance-security §3.1, §5.3 defense

CenTor's anonymity scoring function quantifies the privacy cost of geographic shadow selection using six parameters (client density, AS-level and country-level entropy, relay density, exit density, guard density). Prior work establishes that reducing the client anonymity set by 20x—retaining at least 5% of total Tor users—still provides strong anonymity; accordingly, CenTor recommends minimum thresholds of CD, EL, EC ≥ 0.05 and RD, ED ≥ 0.2 for safe shadow operation.

2025-arora-improving-performance-security §4.1, §4.2, Table 2 defense

Location-aware Tor relay selection algorithms leak client AS-level location: an adversary observing 5 guard selections can guess the client's AS in under 10 attempts, and under 5 attempts after 15 selections. A chosen-destination attack exploiting guard selection bias finds the client's guard relay 94% of the time after 300 destination visits.

2025-lu-rpki-based-location-unaware-tor §2.1 detection

Tor provides substantial and measurable protection against video stream fingerprinting: the best-case FPR at 0.5 recall is 0.0000063 for Tor versus 0.0000008 for HTTPS-only connections, roughly an 8x increase. Translating to world sizes, at 0.5 recall and 0.1 precision the maximum viable platform catalog is 42.9M videos over Tor versus 337.5M over HTTPS-only (Tables 3–4), confirming Tor degrades adversary capability even after an assumed prior website-fingerprinting step that identifies video platform visits.

2025-walsh-improved-open-world-fingerprinting §4.5.1, Tables 3–4 evaluation

The paper concludes with design guidelines for future FIA-based privacy-enhancing technologies, identifying that path-aware routing in SCION and NDN's in-network caching both create new surveillance exposure: SCION path headers reveal routing metadata to on-path censors; NDN caching at routers means content is replicated at points under censor control. The authors recommend that PETs built on FIAs treat these architectural features as threat vectors, not privacy benefits.

2025-wrana-sok-surveillance §6 policy

Wrana et al. systematically assess how well existing surveillance and censorship mechanisms can target users of Future Internet Architectures (FIAs) — including NDN, SCION, XIA, and MobilityFirst — finding that DPI and flow-correlation techniques from the current internet map onto FIA traffic with moderate adaptation. The paper identifies that FIA naming/addressing schemes introduce new censorship attack surfaces (e.g., content-name-based filtering in NDN) not present in IP-based architectures.

2025-wrana-sok-surveillance §4, §5 evaluation

Per-flow RTTdiff detection rates are only ~20% because the majority of proxy flows connect to CDN-cached content (Cloudflare, Google, Fastly) that sits within 5ms of the proxy, suppressing the discrepancy. However, aggregating across flows per website visit yields detection rates exceeding 70%—and from the abstract, approximately 80% of top-5K domains generate at least one detectable flow—with half of those detections made within the first 60 packets. This means an adversary can reliably expose client and proxy IPs after just a few website visits.

2025-xue-discriminative §VI-C-2, Table II detection

IMAP/SSL traffic on port 993 constitutes less than 1% of total ISP traffic but accounts for nearly one third of all false positives in the RTTdiff exploit, because IMAP's non-RESTful multi-connection pattern violates the request-response correlation assumption. The overall per-flow FPR is bounded at 0.6–0.7% (on par with GFW's estimated FPR against fully-encrypted proxies), but implementing a pre-filter to whitelist IMAP traffic reduces the FPR by approximately one third, making the fingerprint substantially more precise.

2025-xue-discriminative §VI-C-3, Table III detection

Cross-layer RTT discrepancy (RTTdiff) is a protocol-agnostic fingerprint that exploits an inherent architectural property of all proxy setups: transport-layer sessions terminate at the proxy while application-layer sessions remain end-to-end. Evaluation across 10 proxy protocols—including VMess, Shadowsocks, VLESS, Trojan, XTLS-Vision, and obfs4-wrapped SOCKS—shows near-identical detection rates for all except obfs4, confirming the fingerprint is not tied to any specific obfuscation scheme. At FPR=0.01, per-website detection rates exceed 70% across all tested client and proxy location combinations.

2025-xue-discriminative §I, §VI-C detection

Against the state-of-the-art DeepCoFFEA flow-correlation attacker, FC-DeTorrent reduces the true positive rate at a 10^-5 false positive rate to approximately 0.12 — less than half that of the next-best defense Decaf (TPR ≈ 0.29) — while using 97.3% bandwidth overhead, without delaying any real traffic packets.

2024-holland-detorrent §4.2, §6, Table 5 evaluation

The root cause of port-shadow vulnerabilities is that connection-tracking frameworks maintain five shared, globally-accessible resources across all VPN clients on the same server. The paper's formal model identifies these as: the conntrack table, the NAT table, the port space, the routing table, and the ARP/neighbor cache. Any of these shared resources can be used as a side-channel. Bounded model checking confirmed that enforcing strict process isolation around all five resources eliminates the attack surface.

2024-mixon-baca-snitch §4 (Formal Model), §6 (Mitigations) detection

The "port shadow" exploit abuses five shared, limited resources in Linux conntrack/Netfilter (and analogous frameworks in BSD, Windows) to let an off-path attacker intercept or redirect encrypted VPN traffic, de-anonymize a VPN peer's source IP, or portscan a peer hidden behind a VPN server — all without compromising the VPN's cryptographic layer. Four concrete attacks are demonstrated; formal model checking with bounded model checking verified six process-isolation mitigations that prevent the shared-resource collision.

2024-mixon-baca-snitch Abstract, §3, §5 detection

DeTorOS enables provable geographic avoidance for Tor onion services by running a TEE-backed Bento function as a trusted middlebox: both the client and the onion service upload their respective 3-hop circuit halves to this enclave, which computes the never-once or never-twice avoidance proof without revealing either party's circuit to the other.

2023-arora-detor-onion §3 defense

Computing a never-once avoidance proof for a 6-hop onion-service circuit takes an average of 64.85 seconds — incurred once at connection setup — because the system must collect round-trip timing measurements across all six relays before running the geographic proof; SGX execution overhead is nominal, and the paper notes that lower-RTT circuits (more likely to be DeTorOS-compliant) reduce subsequent data-transfer latency.

2023-arora-detor-onion §5.3 evaluation

Never-twice provable avoidance succeeds for 72.4% of sampled source-destination pairs on 6-hop onion-service circuits, compared to approximately 98% on the original 3-hop DeTor circuits; the degradation arises because the additional hops increase round-trip time, making it harder to rule out forbidden-region traversal via speed-of-light bounds.

2023-arora-detor-onion §5.2 evaluation

Tor's built-in country-exclusion mechanism is unreliable: circuits configured to exclude US Tor nodes only actually bypassed the US 12% of the time, motivating provably-avoidant circuit construction.

2023-arora-detor-onion §1 evaluation

PushProxy decouples upstream (XOR-obfuscated UDP) from downstream (FCM push notifications), implementing triangular routing that prevents per-flow traffic analysis: a network adversary with limited visibility cannot correlate upload and download flows since they use different transport protocols and paths. Median TTFB was 572ms versus 492ms (Shadowsocks) and 508ms (OpenVPN), while performance remained stable during Chinese peak hours (20:00–02:00 GMT+8) when Shadowsocks download times increased from 3s to over 100s.

2023-xue-use §4, §5.2, Figure 7–8 defense

CRON restricts multi-hop covert circuits (N≥1 relays) to delay-tolerant traffic only, because establishing multiple simultaneous WebRTC video calls is 'highly atypical in normal user profiles' and would trigger S1 behavioral anomaly detection. Real-time interactive tunneling is limited to direct circuits (N=0) within pre-existing calls, and active mode introduces only bounded variability in call times and frequency to stay within plausible user-profile ranges.

2020-barradas-towards §4.1 defense

Even when individual WebRTC flows pass traffic analysis, a censor can identify CRON users via three long-term statistical attack types: S1 (simultaneous video calls, atypical for normal users), S2 (sudden connections to previously unknown parties), and S3 (calls at anomalous times, frequencies, or durations). Relay nodes in multi-hop circuits are particularly exposed via S1 because conducting multiple simultaneous video calls is highly atypical in normal user profiles.

2020-barradas-towards §4.1 detection

Slitheen++ achieves a median covert site loading time of 7 seconds in the naive setup, rising to 8 seconds with crawling and 13 seconds with a 1-second thinking-time (TT) delay. The Baseline-to-Covert factor ranges from 3.7–8.5 without TT and from 7.6–21.4 when crawling and 1-second TT are combined, reflecting the fundamental tradeoff between stealth overt behavior and covert throughput.

2020-birtel-slitheen §5 (Figure 1, Figure 2) evaluation

The original Slitheen appended covert upstream data directly to overt HTTP requests, significantly changing upstream traffic patterns and enabling censor identification even when traffic is encrypted. This upstream traffic analysis vulnerability—absent from Slitheen's original threat model—is the primary weakness Slitheen++ addresses.

2020-birtel-slitheen §1, §4 detection

A censor can identify Slitheen relay connections by observing that all packets in a suspected overt flow arrive in strict order while flows from the same source naturally exhibit out-of-order delivery: the relay station's traffic-server component reorders TCP segments to enable TLS record decryption, creating a statistically anomalous per-connection ordering pattern. The reordering buffer also increases per-packet round-trip times, providing a secondary timing signal.

2020-birtel-slitheen §4, §6 detection

MIMIQ leverages QUIC's connection migration to change a client's IP address mid-connection without disrupting ongoing transfers. QUIC's specification requires that endpoints never reuse connection IDs during migration and that migration tokens are encrypted; this makes migration events indistinguishable from a second distinct client initiating a new connection from a fresh IP address.

2020-govil-mimiq §3.1 defense

A censor using latency analysis to classify decoy routing sessions achieves a maximum F-score that drops to nearly 0 when the base rate of decoy routing falls below 10^-4 (one in 10,000 connections). Even at higher adoption rates the F-score remains below 0.5 for most overt sites, making reliable detection infeasible without unacceptable false-positive rates on legitimate traffic.

2018-bocovich-secure §4.2 evaluation

Decoy routing systems that re-encrypt TLS application data across the relay station (Slitheen, Rebound, Waterfall) are vulnerable to nonce-reuse attacks: an adversary capable of observing traffic on both sides of the relay can exploit reused GCM nonces to decrypt or modify covert traffic. Although this falls outside the standard decoy routing threat model, it poses a concrete risk to users already under heightened surveillance who face adversaries with broad network visibility.

2018-bocovich-secure §5.2 detection

DeTor circuits have significantly lower end-to-end RTTs than standard Tor circuits because high-RTT paths cannot satisfy avoidance proofs, effectively self-selecting for shorter routes. Bandwidth distributions are similar to standard Tor. However, intentional packet-delay defenses proposed for Tor (to defeat timing attacks) would increase effective δ and reduce DeTor proof coverage, creating a tension between delay-based anonymity defenses and RTT-based geographic avoidance.

2017-li-detor §6.2.3, §6.2.1 (δ discussion) evaluation

Never-twice avoidance — ensuring no country appears on both the entry leg (source→entry) and exit leg (exit→destination) of a Tor circuit — succeeds for 98.6% of source-destination pairs not in the same country, using only client-side RTT measurements. This directly defeats traffic-correlation deanonymization attacks that require an adversary on both legs of the circuit simultaneously.

2017-li-detor §4.2, §6.3.1 defense

DeTor proves geographic avoidance using speed-of-light RTT constraints rather than Internet topology maps. If the measured end-to-end RTT satisfies (1+δ)·Re2e < Rmin, where Rmin is the theoretical minimum RTT that would include any point in the forbidden region, then packets provably could not have traversed that region — even against adversaries who forge traceroute and BGP responses.

2017-li-detor §3.1, §4.1 defense

Tor's built-in country-exclusion feature provides only the illusion of control: among circuits configured to exclude the US, only 12% could be identified as definitively avoiding US territory. The remaining 88% of 'trusted' circuits fail to deliver a proof of avoidance, meaning standard Tor policy and provable security diverge sharply.

2017-li-detor §1, §6.2.1 evaluation

CloudTransport's passive-rendezvous design ensures clients never establish direct connections to bridges; consequently, even a censor in complete control of a bridge cannot enumerate client IP addresses without computationally intensive flow-correlation analysis. Blacklisting the IP address of a CloudTransport bridge has zero effect on CloudTransport connections, and when a bridge migrates to a new IP address this change is completely transparent to clients.

2014-brubaker-cloudtransport §4.3 defense

The FNPProbeRequest message, designed to return location and uptime of a node sampled via an 18-hop Metropolis-Hastings random walk, can be used to reliably track individual node online times — capturing >98% of online nodes per sampling interval — enabling intersection attacks on anonymity even though it cannot target a specific node by design.

2014-roos-measuring §5.2 detection

Centralized communication architectures have a single global point of failure: governments can leverage centralization to surveil with or without operator cooperation, as demonstrated by the Snowden revelations about Skype, Facebook, and Google. A compromised broker in a centralized design enables monitoring and censorship that spans all users of the service.

2014-tan-censorship §2.1 detection

Pseudonymity is insufficient for dissent networks: social-network profile information can be correlated with external data to deanonymize users, and fixed-infrastructure networks enable localization attacks even without explicit identity. The authors argue that true anonymity—or at minimum strong deniability where usage is non-incriminating and activity is difficult to trace—is required to protect participants.

2013-hasan-building §5.2 defense

The paper proves that immediate forward security is impossible for Telex-like decoy-routing systems. The Telex station must decide whether to treat a connection as a Telex request after the first client message, using only received messages and its long-term key — an eavesdropper who stores all network traffic can replay the station's entire view once it compromises the station's long-term key, retroactively decrypting all sessions.

2013-ruffing-identity-based Appendix B defense

Traffic analysis poses a concrete throughput ceiling: a conservative SWEET user can perform only 35–70 web downloads per day or 10–20 interactive web sessions while staying within the bounds of normal email volume (2012 averages: 35 sent, 75 received daily). Most websites require fewer than 3 SWEET emails in each direction, with Yahoo as an outlier due to its many hosted objects.

2013-zhou-sweet §5.2 evaluation

A warden can fingerprint the specific covert destination a Telex user is visiting by comparing observed latency distributions against a pre-built database of covert-destination latencies. With an intelligently filtered database of only 10 distributions (K-S inter-entry threshold 0.8), the AUC is 0.868, and with approximately 12 collected samples the false positive rate drops below 10%. Larger databases (size 50) degrade to AUC 0.537 due to distribution similarity, but threshold-based filtering restores substantial discriminative power.

2012-schuchard-routing §5.3, Figures 7, 9 detection

A passive timing attack using the Kolmogorov-Smirnov test on connection latency distributions reliably distinguishes Telex users from honest hosts: K-S scores against the overt destination max out at 0.26, while scores against covert destinations (even those within 10 ms of the Telex station) range from 0.3–1.0 with a median of 0.7 for nearby servers and 1.0 for Alexa top-100 sites. The attack is effective even for clients 50–250 ms from the Telex station, with no K-S score below 0.26 observed across 40 PlanetLab hosts.

2012-schuchard-routing §5.1–5.2, Figures 5–8 detection

COR circuit construction enforces four properties to prevent single-entity de-anonymization in a limited-provider setting: (1) entry and exit ASPs must differ; (2) entry and exit CHPs must differ; (3) the same ASP's relays must not surround another ASP's relay without an intervening hop of a distinct ASP; and (4) at least two relays per traversed datacenter so an adversary with only perimeter visibility cannot trivially correlate ingress/egress.

2011-jones-hiding §2.4 defense

Encrypting traffic at the application layer still discloses communicating parties to every ISP along the path; overlay anonymization is subject to blacklisting of exit nodes and traffic analysis. The paper argues that effective privacy requires building anonymity into the network routing layer itself, with the necessary tradeoff being hardware cost and routing inefficiency for privacy-requiring circuits.

2011-liu-tor §1, §8 defense

Tor-like anonymizing overlays are easily censored because they rely on centralized, publicly visible relay lists; governments can blacklist Tor nodes or monitor all Tor exit traffic so that traffic analysis can reveal the source. Traffic to or from Tor 'essentially advertises itself as probably worth tracking.'

2011-liu-tor §1–§2 detection

Tor bridges that always accept incoming connections enable a three-phase 'bridge aliveness attack': an adversary collects bridge descriptors at scale, correlates bridge uptime timestamps with pseudonymous post timestamps to narrow the candidate set (winnowing), then confirms identity via circuit-clogging and timing attacks. Because bridge descriptors remain valid indefinitely and the BridgeDB rate-limits only to one descriptor set per /24 prefix per week, an adversary with botnet or open-proxy access can hoard enough bridges for the winnowing phase to succeed.

2011-smits-bridgespa §1, §1.1 detection

Adding dummy traffic to any anonymity mechanism yields the corresponding kind of unobservability: 'A mechanism to achieve some kind of anonymity appropriately combined with dummy traffic yields the corresponding kind of unobservability.' DC-nets achieve sender anonymity and MIX-nets achieve relationship anonymity; with dummy traffic both achieve the corresponding sender and relationship unobservability respectively.

2010-pfitzmann-terminology §8 defense

Pseudonymity uses persistent identifiers other than real names, enabling accountability while providing partial unlinkability; however, use of the same pseudonym across different contexts enables linkability: the attacker can link all data related to a pseudonym. Unlinkability of two messages requires that the attacker cannot sufficiently distinguish whether they share a sender or recipient; for a scenario with n senders, this holds iff the probability of common authorship is sufficiently close to 1/n.

2010-pfitzmann-terminology §4, §9, §11 defense

The paper establishes a strict property hierarchy: unobservability ⇒ anonymity, and sender/recipient anonymity ⇒ relationship anonymity. Unobservability is strictly stronger than anonymity because it additionally requires undetectability against all uninvolved subjects — the IOI's very existence must be hidden — while anonymity only hides the subject's relationship to the IOI.

2010-pfitzmann-terminology §7 defense

Cloud locality — building clouds from semantically close peers via short-distance links — ensures that 2-wise and 3-wise cloud intersections have median cardinality between 40 and 50 peers, and the probability that a peer participates in clouds whose pairwise intersection falls below 40 is below 10⁻⁴, rendering intersection attacks infeasible in practice.

2009-backes-anonymity §5.3 defense

The surrounding attack on peer anonymity is also effective only when adversaries control at least 50% of the ~100 semantically closest peers to the target; at 25% malicious peers, at least 10 honest peers still join the target's cloud at every step of the joining algorithm, preserving k-anonymity.

2009-backes-anonymity §5.2 evaluation

A censor hosting Skype supernodes can perform passive traffic-flow analysis on relayed streams even without breaking encryption, since supernode-relayed conversations expose traffic metadata. However, with thousands of supernodes in the Skype network, the probability that any censor-controlled supernode relays a specific SkyF2F tunnel is low, making large-scale correlation high-cost.

2009-cao-skyf2f §V.A detection

A circuit-clogging attack against bridge operators—using median-normalized latency correlations—achieved an AUC of 0.884 and an equal error rate of 0.2 when distinguishing the victim bridge from innocent bridges in PlanetLab experiments with 180 victim and 180 disjoint runs. With 10 repeated clogging experiments and a majority-vote threshold, the false positive (and false negative) rate drops below 0.033, confirming a bridge operator's identity with high confidence given a candidate set of ≤4.4 bridges from the winnowing stage.

2009-mclachlan-risks §3.4 detection

An 'unfair queuing' mechanism that partitions CPU time between bridge-operator circuits and bridge-client circuits using a time-allocation parameter τ=0.9 reduced the circuit-clogging AUC from 0.884 to 0.520 (median-normalized) and 0.412 (mean-normalized)—indistinguishable from random guessing—in 20 PlanetLab experiments. The mechanism eliminates latency interference between the two circuit types without requiring the bridge to ever refuse connections, but introduces up to 1−τ performance loss for client traffic.

2009-mclachlan-risks §4.3 defense

Cross-referencing the online/offline status of 87 monitored bridges against 186,935 Wikipedia users' edit sessions showed that 95.7% of users with 50 or more sessions matched zero bridges after winnowing. For users with 180 or more sessions (a surrogate for long-term pseudonymous activity), only 89 false positives remained among 2,329 users—a false positive rate of 0.000439—meaning that even if 10,000 Tor clients volunteer to bridge, on average only 4.4 bridges remain after the winnowing stage.

2009-mclachlan-risks §3.3 detection

Tor encrypts payload but does not obscure traffic volume, leaving a residual publisher-vs-reader asymmetry: a user publishing a home video generates a markedly different upload/download ratio than one reading news. The paper also notes that website fingerprinting attacks — where the adversary pre-downloads hundreds of popular sites and matches traffic patterns to a Tor client's stream — remain possible even through bridge circuits, and are exacerbated by Tor's varying supported protocols (web vs. IM produce different timing signatures).

2006-dingledine-design §8.2 detection

Anderson establishes that anonymity and physical redundancy are substitutes: 'Anonymity enables us to reduce diversity.' Tamper-resistant hardware security modules controlling anonymized file servers ensure no identifiable group of people — including sysadmins — can locate or delete a specific file without breaking a quorum of hardware modules distributed across jurisdictions.

1996-anderson-eternity §4.5 defense

Traffic analysis is identified as the primary threat to location secrecy in a distributed anonymous storage system: if an adversary can correlate inter-server communications or link requests to stored file locations, it can target physical seizure. The paper proposes mix-nets (Chaum 1981) for user-facing file delivery and dining-cryptographers ring protocols for inter-server communications, supplemented by traffic padding, so that even traffic analysis yields no actionable location information.

1996-anderson-eternity §4.5 detection