decoy-routing   Decoy routing

Iran's censorship of refraction-networking proxies (Conjure via Psiphon) is not monolithic: different ISPs independently deploy different techniques and timelines. Over 800 million logged Conjure connections from July 2023–February 2025 across 10+ Iranian ASes show TCI (AS58224, ~33% of traffic) uses packet injection, while MCCI/Hamrah-e Avval (AS197207, ~22%) applies IP-based blocking, and some ASes (Parsonline AS16322, Shatel AS31549) show no proxy blocking at all.

2025-alaraj-iran-refraction §4 detection

Two Iranian ASes apply a protocol allowlist that drops traffic not matching known application-layer protocol patterns (after ~6 packets), independently of the destination IP. Experiments with fresh /26 phantom subnets showed that prefixing Conjure connections with a plain HTTP GET payload evaded this blocking for four weeks, while TLS Client Hello-prefixed and SSH-prefixed connections were blocked within 72 hours (TLS) or 72 hours after port rotation (SSH). HTTP GET on port 80 was the only tested prefix that survived the full experiment window.

2025-alaraj-iran-refraction §4.4, §5 detection

MCCI (AS197207) blocks proxy IPs proportionally to observed connection volume: the more connections a phantom IP receives, the faster it gets blocked. A controlled experiment with a fresh /27 IPv4 subnet divided into 7 /30 sub-ranges with increasing weights confirmed that higher-weighted subnets were blocked first, demonstrating that the censor infers proxy IP reputation from traffic rate rather than from a static blocklist.

2025-alaraj-iran-refraction §4.2.3 detection

Longitudinal AS topology studies cited by the authors show that 95% of core-to-core AS links remain unchanged year-over-year and that large transit providers adjust their peering only gradually, with almost all churn occurring at the customer edge. This implies that high-usage transit ASes identified for RN proxy deployment are likely to retain their topological position for months to years, lending temporal robustness to placement recommendations derived from a single measurement snapshot.

2025-umesh-improved §5.5 evaluation

Proxy placement requirements vary dramatically by country topology: Turkmenistan requires just 1 AS for 75% coverage, Oman requires 3, Afghanistan 5, Iran 10, and China 12. Turkmenistan's extreme centralization means a single transit AS intercepts virtually all paths, whereas China's fragmented routing fabric demands far more deployment sites to achieve equivalent coverage.

2025-umesh-improved §5.4, Table 4, Figure 6 evaluation

When politically uncooperative ASes are excluded from the candidate pool — specifically Russia's AS12389 and Iranian transit ASes AS49100 and AS198154 — the framework recomputes cumulative coverage over remaining candidates and still identifies viable cooperative deployment sites for Iran. This demonstrates that geopolitical filtering can be incorporated into the placement optimization without losing coverage entirely.

2025-umesh-improved §5.3, Figure 5 deployment

For Iran, a greedy cumulative-coverage analysis over 22,799 resolver-to-uncensored-AS paths shows that the top 5 ASes cover 59% and the top 10 ASes cover 76.6% of all DNS resolution paths. AS3257 (GTT Communications) and AS174 (Cogent Communications) each appear in approximately 15.7% of paths and contribute nearly all their usage as unique (non-overlapping) paths.

2025-umesh-improved §5.2, Figure 3–4 evaluation

An AS+IXP multigraph fusing CAIDA traceroutes (13.6M paths), 256M BGP updates from RouteViews/RIPE RIS, and IXP membership data yields 87,157 AS vertices, 1,588 IXP vertices, and 510,810 edges — an order of magnitude richer than BGP-only baselines. Hidden private peering links and IXP fabric connections invisible to BGP alone materially affect coverage estimates for refraction networking proxy placement.

2025-umesh-improved §1, §4.5 evaluation

Because Oscur0 starts with 0-RTT data lacking a full handshake, the station-side connection establishment is vulnerable to replay attacks. Oscur0 mitigates this by including a random 10-byte nonce in the encrypted application data of the first packet; the station checks each arriving nonce against a bloom filter of recently-seen IDs and drops duplicate connections, preventing replay without requiring a full round-trip handshake.

2024-chen-extended §3 Design defense

Testing from a VPS in Iran showed that standard DTLS handshakes are blocked at that vantage point, but Oscur0 avoids this blocking by transmitting only Application Data packets (with Connection ID extension per RFC 9146) after the initial one-shot setup packet, never completing a visible DTLS handshake. A proof-of-concept was implemented in approximately 600 lines of Go using the pion/dtls library.

2024-chen-extended §3 Design / Implementation evaluation

Oscur0 eliminates Conjure's separate registration phase by steganographically encoding ECDH public key, phantom IP, and transport parameters into the encrypted application data of the first UDP (DTLS 1.2 with Connection ID) packet sent to the phantom IP, using Elligator encoding to make the public key indistinguishable from random bytes. This removes several round trips — registration, TCP handshake, and application handshake — compared to standard Conjure, and means censors cannot block the scheme by blocking registration alone.

2024-chen-extended §3 Design defense

Registration-dependent Refraction Networking schemes such as Conjure create multiple single points of failure: censors can block registration channels independently of phantom connections. Domain fronting, a primary registration channel, has been progressively banned by major CDNs — Microsoft Azure in 2021 and Fastly in early 2024 — reducing its viability as a covert registration mechanism.

2024-chen-extended §1 Introduction detection

Prior circumvention transports that tunneled over VoIP or voice-conferencing software were identifiable to censors by their TCP retransmission fingerprint: real VoIP applications do not retransmit dropped packets in the same way, making the covert channel's reliability mechanisms a distinguishing artifact. DTLS and QUIC avoid this because they natively support both fault-tolerant and sequential delivery modes without external indicators of which mode is active.

2024-chen-extended §2 The Case for UDP detection

CDN infrastructure causes 61%–92% of country-specific Alexa top-1k websites to be hosted within the client's own country across India, Iran, Saudi Arabia, Brazil, and the US, as measured by the authors' R-CBG multilateration technique achieving >89% accuracy. This traffic localization means web requests to popular sites rarely cross national borders, undermining the foundational assumption of decoy routing, domain fronting, CacheBrowser, and CovertCast.

2021-gosain-too §4.2, §5.2 evaluation

Conjure's initial registration step requires the client to connect to an overt website hosted outside the censor's jurisdiction before deriving the unused IP address for actual decoy routing, but CDN traffic localization means this bootstrap connection frequently terminates at a local front-end and never crosses the border. The paper finds that for India's Alexa top-100 sites, only 23 websites had any parallel (leaf) HTTP connections terminating outside the country, with a median of just 3 such external leaf connections per site.

2021-gosain-too §5.2, §6.2 evaluation

OUStralopithecus (OUStral), a Selenium-based OUS implementing empirically-derived human browsing distributions — Weibull dwell times (λ=30s, k=0.75), Von der Weth action probabilities (45.1% internal-link clicks, 33% new-URL navigations), and Dubroy tab-switching rates — generated 471 requests with all Cloudflare Bot Management scores above the recommended blocking threshold of 30, while Slitheen and Waterfall consistently scored 1. Because Cloudflare has full HTTP-layer visibility (unavailable to a passive network censor), the paper argues a censor observing only encrypted traffic would be even less able to flag OUStral.

2021-lorimer-oustralopithecus §3.1, §5.2.1, Figure 6 defense

Prior overt user simulators (OUS) using PhantomJS — including Slitheen, Waterfall, and Slitheen++ — received Cloudflare Bot Management scores of 1 (certainly bot-generated) and would be blocked by any operator following Cloudflare's recommended cut-off of 30. Slitheen++ improved marginally by adding user-agent randomization and brief inter-request pauses, but all PhantomJS-based OUS implementations were trivially detectable as bots.

2021-lorimer-oustralopithecus §5.2.1, Figure 6 detection

Across tunnelling systems that apply traffic shaping against ML adversaries, a clear throughput cost emerges: Slitheen + OUStral with WebM replacement achieves up to 2.2 Mbps with 4.7x overhead; Protozoa (WebRTC, end-to-end) achieves up to 1.4 Mbps; DeltaShaper (VoIP) achieves only 7 kbps at 2x overhead. By contrast, Conjure (no traffic shaping) reaches 100 Mbps. Additionally, end-to-middle decoy-routing deployments incur a throughput penalty from packet-boundary parsing at the relay station that end-to-end systems (Protozoa, DeltaShaper) avoid.

2021-lorimer-oustralopithecus §5.3.2, Table 2 evaluation

Extending Slitheen to replace WebM video/audio frames reduced mean overhead from ~20x (image-only Slitheen) to 4.7x (±1.6) over 100 ten-minute sessions, while raising throughput to a mean of 581.7 kbps in video-only mode (max 2023.3 kbps, min 78.2 kbps) and 721.6 kbps in background-video mode (max 1528 kbps). This compares favorably to DeltaShaper's 2x overhead at only 7 kbps and Protozoa's up to 1.4 Mbps, while preserving Slitheen's resistance to traffic-analysis attacks.

2021-lorimer-oustralopithecus §5.2.2, §5.3, Table 2, Figures 7–9 evaluation

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

Slitheen++ embeds covert upstream data by applying HTTP/2-like header field compression to overt HTTP requests, using the recovered space for covert data placement. This ensures that neither timing information nor observable changes to packet sizes or delays can reveal decoy routing use to an omni-scientist passive censor. GZIP compression was explicitly avoided to prevent the CRIME side-channel attack.

2020-birtel-slitheen §4 defense

Slitheen++'s relay station introduces minimal overt forwarding overhead: 95% of setups saw downstream per-packet delays between 1 ms and a maximum of 4 ms, with on average only 0.0029% of downstream packets affected (peak 0.006% in any single scenario). Upstream delays were similarly low except for a single outlier near 60 ms caused by thread contention during crawling-induced relay load spikes.

2020-birtel-slitheen §5 (Table 1) 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

Total hardware cost for four detector stations plus a central proxy was approximately $30,000 USD, with estimated annual operating costs of ~$13,000 for co-location and ~$24,000 for a 2 Gbps upstream connection, plus ~40% FTE of an ISP network engineer—costs that are structurally higher than endpoint-based circumvention due to mandatory network-operator co-location requirements.

2020-vandersloot-running §5.2 deployment

During a major censorship event in April 2019, new censor techniques blocked many Psiphon transports while TapDance remained accessible, causing a 4× increase in the fraction of TapDance-enabled clients' traffic and daily users peaking above 25,000—with no measurable degradation in connection success rate or per-session throughput under the increased load.

2020-vandersloot-running §4.5 evaluation

Over 115 days of operation with 1,500–2,000 active decoys, the busiest decoy site averaged only 13.24 concurrent connections and 12.32 MB of traffic per day; only 2 of roughly 3,000 candidate decoy sites opted out via robots.txt over 18 months, and none reported operational problems, confirming that decoy websites are not meaningfully burdened at this deployment scale.

2020-vandersloot-running §4.3, Table 1 evaluation

The first production Refraction Networking deployment used four TapDance stations at Merit Network observing 140 Gbps aggregate capacity and served up to 33,000 unique users per month across 559,000 Psiphon installations, proxying up to 500 Mbps of circumvention traffic during the first year of continuous operation.

2020-vandersloot-running §1, §3, §4 deployment

Approximately 50% of TapDance client connection attempts failed to pass through any station due to routing variability at the ISP, producing an average of about one failed decoy per successful connection and time-to-first-byte exceeding five seconds in typical cases, even though median session RTT remained under one second.

2020-vandersloot-running §4.4, §5.4 evaluation

Conjure achieves 20% lower latency, 14% faster download bandwidth, and over 1400 times faster upload bandwidth compared to TapDance on a 20 Gbps ISP testbed. TapDance upload is throttled to approximately 0.1 Mbps because it must reconnect for every 32 KBytes sent; Conjure maintains a single persistent connection. At the 99th percentile, Conjure is 281 ms (92%) faster than TapDance.

2019-frolov-conjure §6.1 evaluation

For IPv4, Conjure derives both the phantom host IP and TCP port from the client's registration seed, making exhaustive scanning infeasible: a censor enumerating from a /10 of potential client source IPs (4 million addresses) against a /16 of phantom IPs (65K addresses) across all 65K ports would require approximately 50 years at 10 Gbps with ZMap. Phantom hosts are additionally firewalled to respond only to the registering client IP, defeating single-vantage-point ZMap scans.

2019-frolov-conjure §6.2.1 defense

IPv6 phantom addresses drawn from an ISP's /32 prefix provide 2^96 potential addresses, making exhaustive enumeration and pre-image attacks computationally infeasible. Analysis of 4013 observed IPv6 addresses in a deployed /32 found approximately 75 bits of entropy (out of a maximum 96), with enough overlap with legitimate address distributions that blocking high-entropy addresses would produce significant collateral damage to real IPv6 services.

2019-frolov-conjure §6.2.2 defense

Conjure phantom hosts resist active probing by requiring knowledge of a per-client registration seed secret before the station responds. A ZMap scan of over 1 billion random IP/port combinations found that 99.4% of responding servers returned no data after a random OSSH-style probe and 7.42% closed with TCP RST — behavior indistinguishable from Conjure's OSSH transport — meaning censors face steep false-positive rates when attempting to identify phantom proxies via active probing.

2019-frolov-conjure §7.1 defense

Conjure registration is unidirectional: the client embeds a steganographic ciphertext tag in a complete HTTPS request payload encrypted under a Diffie-Hellman shared secret, and the station passively observes it without sending any reply or spoofing packets. This design makes registration flows indistinguishable from normal HTTPS traffic and enables 25% more viable registration decoys than TapDance by removing the requirement to exclude decoys with short TCP windows or connection timeouts.

2019-frolov-conjure §4.1 defense

For China (a highly connected, routing-capable adversary), the gossip protocol combined with any symmetric decoy routing design requires only 5 heavyweight downstream stations plus 880 lightweight upstream gossip stations — versus 880 heavyweight stations for purely symmetric designs. Five downstream stations alone impact 78% of routes from Chinese users, while a single downstream station already covers nearly 25% of traffic.

2018-bocovich-secure Table 2 / §3.2 evaluation

An asymmetric gossip protocol adds only 1.0055× bandwidth overhead for n=5 downstream stations — approximately 11 Mb/s extra on a typical 2 Gb/s OC48 link. Upstream gossip stations require no in-line blocking and impose zero additional load on overt sites, making them substantially lighter than heavyweight symmetric relay stations that must check every TLS connection for steganographic tags.

2018-bocovich-secure §3.3 evaluation

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

Between 80% and 90% of internet routes are asymmetric, with only about 10% of flows symmetric in Tier-1 (backbone) networks and roughly 70% symmetric at the network edge. This asymmetry makes decoy routing systems requiring relay stations on both upstream and downstream paths impractical for the majority of real-world deployments.

2018-bocovich-secure §3 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

MultiFlow's tunnel operates as a virtual message board: the client and decoy router never exchange covert data within the same TCP connection. The decoy router uploads responses to a URI or email address specified by the client; the client downloads independently on a separate connection. This design eliminates the forged-packet and rewritten-traffic vectors that make TapDance and Rebound vulnerable to traffic analysis and decoy-host probing.

2018-manfredi-multiflow §3.2 defense

MultiFlow mitigates TLS termination attacks—where an adversary drops a connection after one data exchange—by having the client exfiltrate TLS session resumption information (219 bytes: 208-byte psk identity plus ticket metadata) to the decoy router. The decoy router can then resume a session with a different decoy host, establishing a new covert channel even if the original connection is severed, and amortizing per-session setup cost across multiple connections.

2018-manfredi-multiflow §3.3 defense

If an adversary replays captured client handshake traffic to a decoy host under adversary control, and the decoy router attempts to resume the client's session on that host, the adversary can infer that a decoy router is present on the path to the original decoy host. The paper identifies this as a residual probing vulnerability when the client does not encrypt the destination server to which resumption should be directed.

2018-manfredi-multiflow §3.1 detection

MultiFlow's stencil-coding capacity is constrained by TLS record sizes: hiding 1 byte per 16-byte block requires a 1568-byte TLS record to exfiltrate 98 bytes of key material. The paper notes that many websites' initial GET requests produce TLS 1.3 application records under 100 bytes, meaning MultiFlow would need to span multiple records or adopt the more efficient chosen-ciphertext steganography used by TapDance. No implementation exists at time of publication; session resumption from a different source IP was verified feasible using OpenSSL 1.1.1-pre2 and Scapy.

2018-manfredi-multiflow §3.1, §3.4 evaluation

MultiFlow enables a tap-based decoy router to authenticate clients without inline traffic blocking by having the decoy router resume the client's TLS 1.3 session with the decoy host. The client embeds 112-byte sentinel values in the ClientRandom and key-share fields; the decoy router uses the exfiltrated 219-byte NewSessionTicket to perform the resumption. If the decoy host accepts the resumed session rather than falling back to a full handshake, the client is confirmed live.

2018-manfredi-multiflow §3.1 defense

Without per-site connection limits, popular decoy hosts risk resource exhaustion (Apache's default cap is 150 simultaneous connections); enforcing an initial limit of 30 concurrent clients per site—coordinated across stations via a central collector—kept the median site load at ~5 simultaneous clients, with the 99th-percentile site peaking at 37 after the limit was raised to 45.

2017-frolov-isp-scale §3.5, §4.3, Figure 9 deployment

Filtering candidate decoy sites by a minimum 15 KB TCP window eliminated 24% of the initial ~5,500 HTTPS hosts; a 30-second HTTP-timeout floor eliminated a further 11%; and AES-128-GCM cipher-suite support requirements eliminated an average of 32%—together reducing the viable decoy-site pool by approximately 55% before any live reachability tests.

2017-frolov-isp-scale §3.3, Figures 2–3 defense

The one-week trial served over 50,000 unique users (peak daily count: 57,000) with up to 4,000 concurrent sessions simultaneously, demonstrating that a four-station refraction deployment co-located at two mid-sized network operators can support tens of thousands of real censored users.

2017-frolov-isp-scale §4.2, Figures 4, 7 evaluation

The trial explicitly obtained no evidence about TapDance's resistance to adversarial censor countermeasures: its scale and duration were judged small enough that censors likely did not observe it, leaving theoretical censorship-resistance claims unvalidated against active blocking responses.

2017-frolov-isp-scale §2, §6 evaluation

TapDance was deployed on four ISP uplinks (two 40 Gbps, two 10 Gbps) using commodity 1U servers running a Rust/PF_RING zero-copy implementation; CPU load remained below 25% while handling a peak of ~14,000 new TLS connections per second across 34 cores, with cumulative mirrored traffic peaking at 55 Gbps across all stations.

2017-frolov-isp-scale §3.1, §4.1, Figure 1 deployment

The 30 key ASes computed from globally popular sites also intercept over 90% of paths to country-specific popular sites in nine censorious nations (China, Venezuela, Russia, Syria, Bahrain, Pakistan, Saudi Arabia, Egypt, Iran), covering 93.3% of paths to the top-50 country-specific sites. The same key AS set remained stable across repeated experiments conducted four months apart, suggesting durability over time.

2017-gosain-devil-s §6.1–6.2, Figure 8 evaluation

Only ~30 ASes intercept more than 90% of paths to popular websites globally, regardless of the target destination set (Alexa top-10 through top-200). The top 2 ASes alone (AS3356 Level-3 Communications and AS174 Cogent) intercept 45.1% of all 4,497,547 paths to Alexa top-100 sites; the full set of 30 intercepts 92.4%. This is approximately 30× fewer ASes than prior work required for a single adversary country (858 ASes for China alone).

2017-gosain-devil-s §5.1, Figure 3, Table 1 evaluation

If China attempts the Routing-Around-Decoys (RAD) attack by blackholing paths that transit the 30 key ASes, 92.25% of all paths transiting Chinese ASes (306,874 of 332,742) originate at ASes outside China, making such filtering self-defeating through severe collateral damage to foreign transit customers. The 30 key ASes cover 98.8% of paths from Chinese ASes to globally popular destinations and at least 80% for nearly all adversary countries studied.

2017-gosain-devil-s §6.2, Figure 7, Figure 9 defense

Customer-cone size — the AS selection metric used by prior work (Houmansadr et al. 2014) — is poorly correlated with actual path frequency (Spearman rank correlation = 0.2). 33.17% of paths to Alexa top-100 prefixes traverse 1-hop customers of the largest-cone AS (AS3356, cone size 24,553) without transiting AS3356 itself, showing that cone-based heuristics systematically misidentify which ASes actually carry traffic.

2017-gosain-devil-s §6.6, Figure 10, Table 5 evaluation

Router-level mapping of the 30 key ASes reveals that 11,709 individual routers must be replaced with Decoy Routers (non-censorious ASes only), at a hardware cost exceeding $10.3 billion USD. Individual large ASes require hundreds to over 1,600 router replacements (e.g., AS3356 needs 576, AS209 Quest Communications needs 1,662). Even targeting the weakest adversary studied, Syria (containable by 3 ASes at AS level), requires 1,117 DRs.

2017-gosain-devil-s §5.2, §6.4, Table 3–4 evaluation

Through Internet-scale BGP simulation against China, downstream-only decoy routing (Waterfall) with a single decoy AS provides equivalent resistance to routing attacks as a traditional upstream decoy system (e.g., Telex) with 53 decoy ASes. This efficiency gain arises because censoring ISPs can selectively re-route upstream traffic per destination but must re-route all or none of downstream traffic through each provider AS, making downstream-only routing far more expensive to evade.

2017-nasr-waterfall §4.1.1 defense

Evaluation of the top 10,000 Alexa websites finds that 3,916 (39%) support HTTPS, of which 1,976 (50%) perform HTTP 3XX redirects that echo the requested path in the Location header and 812 (20%) replay the URL in HTTP 404 error responses — both usable as upstream covert channels readable by downstream-only decoy routers without intercepting upstream traffic.

2017-nasr-waterfall §7.1.1 defense

Waterfall's Overt User Simulator caches previously loaded overt-website responses and replays them to generate cover traffic, overcoming Slitheen's 40% downstream throughput ceiling (caused by restricting covert replacement to leaf HTTP objects only). Because downstream-only decoy routers intercept all downstream TLS records — not just leaf content — Waterfall achieves higher covert capacity while perfectly mimicking overt browsing patterns against traffic analysis.

2017-nasr-waterfall §9 defense

Table 2 shows that with 50 decoy ASes, the most powerful practical routing attack on downstream-only systems (rewiring-I) impacts 93% of China's routes (22.4% unreachable, 70% re-routed), compared to only 18.2% total impact from RAD on traditional upstream designs. Table 3 shows that even for Syria, the rewiring-II attack with just 1 downstream-only decoy AS already impacts 81% of routes versus 1.5% for RAD on upstream systems.

2017-nasr-waterfall §4.2.1, Table 2, Table 3 evaluation

BGP simulation shows that a censor's source-block attack against 100 downstream-only decoy ASes disconnects 23% of Chinese Internet destinations, versus only 8% when applying the standard RAD attack against 100 upstream decoy ASes — imposing nearly 3× more unreachability collateral damage on the censor for the same decoy count.

2017-nasr-waterfall §4.1.1, Figure 3 evaluation

Slitheen replaces only 'leaf' HTTP resources (images, video) in overt-site responses with covert content, reusing all TCP/IP headers verbatim and forwarding packets immediately on arrival. This forces every observable feature—packet size, direction, inter-arrival timing—to be identical to a genuine access of the overt page, eliminating the censor's ability to apply latency analysis, website fingerprinting, or protocol fingerprinting to distinguish decoy sessions from normal traffic.

2016-bocovich-slitheen §3.1, §4.1 defense

Measurement of the Alexa top 10,000 TLS sites showed that the fraction of traffic replaceable by a Slitheen relay varies from 0% (Facebook, due to large TLS records preventing leaf replacement) to 100% (Wikipedia, Yahoo). For representative sites: Reddit achieved 70% ±10% of leaf bytes replaced (19% ±3% of total page bytes), Gmail 87.7% ±0.2% of leaf bytes (23% ±9% total), and Quora 99% ±5% of leaf bytes (20% ±10% total), as reported in Table 2.

2016-bocovich-slitheen §6.1, Table 2 evaluation

Table 1 shows Slitheen is the first decoy routing system to simultaneously defend against latency analysis, website fingerprinting, and protocol fingerprinting attacks, while also resisting TCP replay and Crazy Ivan active attacks. This security is achieved at the cost of requiring symmetric flows and inline blocking—requirements previously considered prohibitive—which the authors argue are increasingly met by commercial DPI traffic-shaping appliances (e.g., Sandvine) already deployed by ISPs.

2016-bocovich-slitheen §4.5, Table 1 evaluation

Internet connectivity is the primary determinant of RAD attack strength across nation-state censors: China (573 ASes, 858 ring ASes) achieves a censorship metric of 0.277 under profile T1, while Syria (4 ASes, 5 ring ASes) achieves only 0.101 with the same decoy budget. Venezuela, despite fewer total ASes than Saudi Arabia (44 vs. 107), achieves a higher censorship metric (0.210 vs. 0.197) owing to its disproportionately large ring AS count (835 vs. 176), confirming that ring AS count predicts RAD capability better than raw AS count.

2016-nasr-game §6.4 / Table 6 evaluation

Optimal RAD by a QoS-cautious wealthy Chinese censor (profile T1, F/ρ₀ = 5×10^6) forces 10.8% of routes onto non-valley-free (NVF) paths and 1.2% onto less-preferred routes, while still leaving 16.3% of routes traversing decoy ASes—zero routes become unreachable at this budget. The NVF and less-preferred-route fractions rise and then fall as decoy budget increases, as further RAD routing gains diminish past a crossover point.

2016-nasr-game §6.4 / Figure 1 detection

The game-theoretic optimal decoy placement (ε-Nash equilibrium via best-response dynamics against an optimal RAD adversary) achieves a censorship metric of 0.2 at budget ratio F/ρ₀ = 10^8, versus 0.42 for the best prior heuristic ('sorted' placement from Houmansadr et al. [14]) under the same budget—a 2× improvement in censorship resistance per dollar. Prior comparisons used ad hoc RAD deployments rather than the optimal adversary, understating the benefit of principled placement.

2016-nasr-game §6.4 / Figure 3 evaluation

Game-theoretic simulation shows that a QoS-cautious, wealthy Chinese censor (profile T1/T4) cannot reduce decoy-accessible routes below ~27% (censorship metric ≈ 0.277) via the RAD attack regardless of budget. An irrational censor can achieve a censorship metric of 1.000 but only by making 70.3% of all Internet routes unreachable to Chinese users—a collateral-damage threshold that constrains rational nation-state censors in practice.

2016-nasr-game §6.4 / Table 5 defense

In the autonomous (non-centrally-funded) deployment model, the decoy service fee γ (ratio of decoy revenue to transit revenue per MB) is the primary lever for censorship resistance: for China with profile T1, γ = 5 leaves 9.6% of routes usable for circumvention after optimal RAD, compared to 16.3% under the centrally-funded model at budget ratio F/ρ₀ = 5×10^6. Higher fees compensate ASes for RAD-induced transit revenue loss and sustain participation, but the autonomous model delivers roughly half the censorship resistance of a centrally-funded deployment at comparable incentive levels.

2016-nasr-game §6.5 / Figure 5 defense

He et al. found that 65% of sampled routes between public traceroute servers have some degree of AS-level asymmetry; John et al. found that asymmetry reaches 96% on Tier-1 ISP backbone links due to hot-potato routing. These figures invalidate the symmetric-route assumption underlying Telex and Cirripede and motivate a fully asymmetric design.

2015-ellard-rebound §I evaluation

Because Rebound never terminates the client–decoy connection, connection-state probes (including 0trace-style TTL-expiry probes that bypass the decoy router via an alternate route) cannot reveal any discrepancy between the observed and actual state: the connection to the decoy host is always exactly in the state a censor would expect.

2015-ellard-rebound §VIII-A2 defense

Rebound's mole protocol generates a characteristic traffic pattern — a steady stream of long HTTP GET requests followed by 404-style error responses — that may be identifiable via traffic analysis even though the channel is TLS-encrypted; the paper acknowledges this as an unmitigated vulnerability and notes that intermingling with ordinary requests reduces observability but further lowers effective throughput.

2015-ellard-rebound §VIII-B detection

Rebound eliminates the stack-fingerprinting vulnerability present in Telex, Curveball, Cirripede, and TapDance by never forging packets addressed to the client; all data from the decoy router to the client travels through the real decoy host, so the TCP/IP stack fingerprint observed by a censor is always that of the genuine decoy.

2015-ellard-rebound §VIII-A1 defense

In an Internet measurement from a residential Verizon FiOS client 12 hops from the Rebound router (26 ms RTT), Rebound achieves 129,398 bytes/s (≈126 KB/s) for 1 MB transfers, compared to 354,676 bytes/s for Curveball and 1,174,240 bytes/s for plain HTTP — sufficient to stream 360p video but roughly 3× slower than Curveball. The unoptimised Python router implementation uses less than half a core of an Intel Xeon E5620 at 2.4 GHz at sustained full speed.

2015-ellard-rebound §VII, Table I evaluation

Under the RAD attack a large fraction of China's routes to Internet destinations shift to non-valley-free (NVF) paths, which impose direct monetary costs because ASes must pay for traffic they would normally earn revenue transiting. Among valley-free paths that survive, 6%–21% switch to less-preferred (more expensive) routes, 20%–43% become longer, and average path length increases by 1.12×–1.40× depending on placement strategy.

2014-houmansadr-no §VII-B, §VII-C evaluation

Even under the most censor-favorable (random-no-ring-1) decoy placement, launching the RAD attack increases average Internet route latency from China by over 4×; under strategic placements the average latency increase factor reaches 8×. These increases arise because RBGP is forced onto lower-capacity, less-popular transit ASes even when path hop-count is unchanged.

2014-houmansadr-no §VII-C, §I evaluation

The feasibility of the RAD attack scales sharply with the censor's network connectivity. Strategic placement of decoys in just 1% of ASes disconnects China from 18% of Internet destinations, Venezuela from 54%, and Syria from 87%. Countries with fewer controlled ASes and ring ASes have dramatically less routing flexibility and are far more vulnerable to even small decoy deployments.

2014-houmansadr-no §I, §VII evaluation

The RAD attack requires converting a large number of Chinese edge ASes into transit ASes: placing decoys in 2% of global ASes (random-no-ring-1, China-World scenario) forces 59 edge ASes to become transit ASes, nearly doubling China's 30 existing transit ASes. One Chinese transit AS must carry approximately 122× its normal load; the abstract reports a peak of 2,800× in a more aggressive scenario, a threshold the paper considers operationally infeasible.

2014-houmansadr-no §VII-D evaluation

The RAD paper's random decoy placement is heavily biased in favor of the censor: 86.2% of all Internet ASes are edge ASes with customer cone size 1, so random selection rarely hits transit ASes. Replacing random with sorted-no-ring placement (decoys chosen from ASes that appear most on adversary BGP routes) disconnects China from 30% of Internet destinations using only 2% decoy coverage, versus the 4% disconnection reported in the original RAD paper.

2014-houmansadr-no §V, §VII-A defense

Known attacks on existing circumvention tools include steganographic detection, enumeration of decoy-router locations, and machine-learning traffic classifiers. The paper acknowledges these defeat current approaches (Infranet, Collage, Telex, SkypeMorph, Freewave) and argues that no iterative patch can neutralize the censor's long-term structural advantage.

2014-tan-censorship §1 defense

All three prior end-to-middle (E2M) schemes — Telex, Cirripede, and Decoy Routing — require an inline flow-blocking component at the participating ISP, which adds latency, introduces a single point of failure, and may violate carrier SLAs. In private discussions with ISPs, the authors found that despite willingness to assist Internet freedom technically and financially, none were willing to deploy existing E2M technologies due to these operational impacts. TapDance removes the inline blocking requirement entirely, requiring only a passive tap and packet-injection capability.

2014-wustrow-tapdance §1, §6 deployment

Replacing Telex's original stego-tagging with the IBST scheme and using time periods as identities achieves eventual forward security with arbitrarily short rotation intervals. The key material a client needs after a master-key rotation is only the new master public key — 'a few hundred bytes' — small enough to fit in covert channels such as steganographic images, avoiding the original Telex design's problem of large bundled key sets expiring before a client updates its software.

2013-ruffing-identity-based §4.2 Application to Telex 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

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 routing-capable warden can enumerate over 90% of decoy-router-deploying ASes for deployments as large as 4,000 ASes using an intersection-based discovery attack: the warden probes many paths, accumulates a set of 'clean' ASes, and prunes candidate paths until a single 'tainted' AS remains. All evaluated wardens (China, Syria, Iran, Australia, France, Venezuela) achieved roughly equal detection success across all deployment sizes.

2012-schuchard-routing §4.1 detection

Containment analysis shows that surrounding China with a 'ring' of decoy routers at AS-hop depth 1 requires covering 161 ASes; depth-2 expands by a factor of more than 23, becoming untenable, while depth-3 is slightly smaller but leaves the majority of the Internet reachable via clean paths. Cutting China off from at least half the Internet would require all 96 of the world's largest ISPs to deploy decoy routers at all exit points simultaneously.

2012-schuchard-routing §6, Table 2 evaluation

Under the Cirripede 'random ASes' deployment scenario — where 0.4%–1.0% of ASes deploy decoy routers — routing-capable wardens need only disconnect themselves from 0.85%–3.04% of the Internet to obtain clean (decoy-free) paths to all remaining destinations. Even at 10% Internet-wide deployment, wardens are cut off from only 7%–9% of non-participating ASes on average.

2012-schuchard-routing §4.2, Figure 2 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

If clients probe the top 1,000 Alexa-ranked sites to discover a deflecting router, a censor would have to block more than 95% of those 1,000 sites to prevent any client from joining Cirripede. Clients aware of failed probes can continue cycling through additional popular sites, further raising the blocking cost.

2011-houmansadr-cirripede §6.3, Figure 5c defense

In an emulation testbed with 200 ms effective client-server RTT, Cirripede added no more than a few seconds to time-to-first-byte, attributable primarily to two extra TLS round-trips and the SOCKS request-response. For large file downloads, Cirripede's TCP connection splitting (two lower-RTT hops instead of one high-RTT hop) produced faster total transfer times than the non-Cirripede baseline, confirmed with a control non-Cirripede SOCKS proxy.

2011-houmansadr-cirripede §6.2, Figure 3 evaluation

Replaying 94 million TCP SYN packets from 6.4 million unique client IPs at ~41,000 packets/second, the Cirripede registration server (quad-core Xeon E5530, 12 GB RAM) achieved a 97% detection rate — 1,038,689 out of 1,069,318 embedded registrations — with average CPU utilization of 56% (max 73%) and average memory of 1.1 GB (max 1.6 GB). The 3% miss rate was caused entirely by network-layer packet reordering, not server capacity.

2011-houmansadr-cirripede §6.1 evaluation

Using two CAIDA traces from March 2011, the byte volume of TCP SYN packets across all ports was only 4–7% that of port-443 traffic. Cirripede's registration design inspects only SYN packet headers rather than full HTTPS payloads, reducing the traffic an ISP must process by 14–25× compared to Telex/Decoy routing architectures that must reconstruct all port-443 TCP sessions.

2011-houmansadr-cirripede §7.2 evaluation

Simulations on the CAIDA AS-level topology (January 2011 snapshot) show that deploying Cirripede deflecting routers at just 1 tier-1 AS enables 97% of Internet clients to use the system, and 2 participating tier-1 ASes achieve 100% client reachability. When clients probe only the Alexa top-30 most popular sites as overt destinations, 2 tier-1 ISPs still yield 100% reachability.

2011-houmansadr-cirripede §6.3, Figure 5b deployment

Decoy routing places the circumvention service at transit routers rather than fixed-IP edge proxies, so the client addresses packets to any reachable decoy destination and the router hijacks the flow on the client's behalf. A single well-placed router may lie on paths to millions of destinations, making circumvention proxies appear ubiquitously deployed from an adversary's perspective. Blocking such a router requires disrupting ordinary traffic for large fractions of the Internet, qualitatively raising the cost of IP-address-based censorship.

2011-karlin-decoy §1.2 defense

An adversary aware of a decoy router's location can force decoy-routed flows to be unprocessable by fragmenting all packets below the size of a complete TCP header in the first fragment, preventing flow assignment and forcing the router into expensive reassembly. Alternatively, the adversary can use small-fragment attacks to grow the router's state table, analogous to NAT resource exhaustion. The paper identifies fragmentation-based denial as a harder-to-mitigate attack class than sentinel replay.

2011-karlin-decoy §4.2 detection

A preplay attack defeats the TLS-sentinel covert channel: the adversary intercepts each ClientHello, immediately sends a copy to the decoy destination before the client's copy arrives, causing the sentinel to be consumed and poisoned. The client can never establish a decoy routing session while ordinary TLS to the decoy destination continues to work normally, giving the adversary both blocking capability and forensic confirmation that decoy routing was attempted. The paper notes this vulnerability is specific to the TLS sentinel and that alternatives such as port-knocking sentinels may not share it.

2011-karlin-decoy §4.1 detection

TCP flow hijacking by the decoy proxy is practical under an asymmetric routing assumption: expected sequence numbers are recoverable from ACK values in client-originated packets alone, so the decoy router need not observe return traffic. The proxy forges a TCP RST to the decoy destination and mimics its TCP options (timestamp, window scale, SACK) to reduce detectability; these options are conveyed encrypted inside the sentinel's 28-byte TLS random field.

2011-karlin-decoy §3.3 defense

Clients embed HMAC-derived, time-varying sentinels into the 28-byte random field of the TLS ClientHello message, which decoy routers can scan at line rate. Sentinels are keyed to the current hour and a per-hour sequence number, providing freshness. This covert channel requires no out-of-band signaling and is invisible to passive observers who see only a normal TLS handshake toward the decoy destination.

2011-karlin-decoy §3.2 defense

On a single 2.93 GHz Intel Core 2 Duo CPU core, the Telex elliptic-curve tagging scheme achieves approximately 5,482 tag generations per second and 11,074 tag verifications per second across 10 trials of 100,000 tags each (standard deviations of 0.016 s and 0.0083 s respectively). Tag verification is therefore unlikely to be a throughput bottleneck in a deployed Telex station.

2011-wustrow-telex §8.2 evaluation

Telex embeds steganographic tags in TLS ClientHello nonces using elliptic-curve Diffie-Hellman, placing proxy stations at ISP level on paths between the censor's network and popular uncensored destinations. Because the cover destinations are ordinary popular HTTPS websites, the censor cannot block Telex without simultaneously blocking a large class of legitimate TLS traffic — converting the censor's own reluctance to over-block into an unblockability guarantee.

2011-wustrow-telex §2, §4 defense

A PlanetLab node in Beijing successfully loaded all 100 Alexa top-100 websites through a prototype Telex station at the University of Michigan; without Telex, 17 of the 100 sites were blocked (including facebook.com, youtube.com, blogspot.com, and twitter.com from the top 10), using forged RST packets, false DNS results, and destination IP blackholes. The median latency overhead for routing through Telex was approximately 60% for the 83 unblocked sites.

2011-wustrow-telex §8.4 evaluation

Telex prevents tag replay attacks by seeding the client's TLS key exchange randomness (e.g., the Diffie-Hellman exponent) with the shared secret ksh derived from the steganographic tag. The TLS Finished message must then be freshly encrypted with the newly negotiated master secret, implicitly proving knowledge of ksh. An adversary replaying a captured ClientHello nonce without knowing ksh cannot produce a valid Finished message, causing the server to terminate the connection.

2011-wustrow-telex §6.2 defense