meta-resistance   Meta-circumvention / framework

Fountain codes (rateless erasure codes) break a message into many small, specially encoded pieces that can be reconstructed from any sufficient subset, making them well-suited to covert channels that are lossy, unordered, or bandwidth-constrained — properties common to DNS-based rendezvous channels.

2026-cawthon-fountain-codes-censorship Abstract defense

Drawing on the Collage and Assemblage censorship-circumvention systems — which use fountain codes in conjunction with steganography — this work narrows focus to the rendezvous step alone, treating fountain codes as a standalone design component useful independently of steganographic cover channels.

2026-cawthon-fountain-codes-censorship Abstract defense

Under April 2026 enforcement pressure, surviving VPN resellers converged on three strategies: raising prices to cover higher infrastructure costs, switching from transit to direct-connect (higher latency, worse peak-hour performance), or deploying proprietary protocols with dedicated clients — the last option breaking compatibility with standard Clash and Shadowrocket clients and fragmenting the interoperable ecosystem.

2026-ermao-april-airport-outage §5月以后机场可能怎么-活下去 deployment

Operators facing the April 2026 enforcement wave described three survival paths: (1) price increases to pass on resource costs, (2) switching entirely to direct-connect, or (3) deploying proprietary protocols with dedicated clients — making standard Clash/Shadowrocket clients non-functional for those providers. The commercial forecast is for low-price high-quality plans to disappear and for month-by-month billing to become the default as users hedge against provider collapse.

2026-ermao-april-airport-outage §5月以后机场可能怎么活下去 deployment

Huma separates proxy duties between untrusted Decoy Websites (DWs), which relay encrypted messages and serve content, and trusted Shade Proxies (SPs) outside the censored region, which decrypt requests and contact covert destinations. Even if a DW is compromised, the censor learns only whether a specific UID can access the system — no destination, no content, and no client network-layer information. SP assignment is centrally managed by the Huma Authority, preventing DW-SP collusion.

2026-kamali-huma §III-A, §IV-C defense

DPYProxy-DNS's automated probe-and-select mode identified a working DNS circumvention in an average of 13.78 seconds (median 12.90s) in China and 12.32 seconds (median 8.28s) in Iran across 100 runs each; best-case startup was 0.32s (China) and 0.47s (Iran) when the first-tried combination succeeded, while worst-case exceeded 30.72s in China and 58.16s in Iran due to the slow Last Response mode (3s fixed wait per attempt) being selected early in the randomized probe order.

2026-lange-towards §6.3 evaluation

Correlated domain burns (e.g., registrar takedowns) collapse the sustainability frontier far more than mean burn rate alone: the interval frontier β* falls from 0.65 at burst size b=1 to 0.5 at b=2, 0.3 at b=4, and at b=8 no β in the tested range is sustainable. Conversely, diversifying the same total domain economy across P=8 independent providers (where each provider's pool can be emptied by a single takedown but providers fail independently) moves the frontier past β=1, and simulated P-provider availability matches the closed form A_P = 1 - q^P geometrically (Proposition 4, Figure 5d).

2026-maiti-block-a-mole-sustainability-frontier §8.7, §8.8, Proposition 4, Figure 5 defense

When the domain burn rate β = λdisc/λintro exceeds 1, time-average availability is bounded by A ≤ 1/β regardless of rotation speed µ, endpoint count n, or domain buffer kmax (Theorem 2). Rotation speed µ appears only in the address factor, which redundancy already saturates, and is entirely absent from the binding domain layer — proving that 'rotate faster' is categorically the wrong design lever under a domain-filtering censor.

2026-maiti-block-a-mole-sustainability-frontier §6, Theorem 2 evaluation

In April 2024 the GFW began decrypting QUIC Initial packets at scale and blocking connections by domain rather than by IP address, targeting the exact transport layer that modern moving-target circumvention systems rely on for session continuity. Because blocking a registrable domain costs the censor essentially zero collateral (unlike blocking a hyperscaler IP range), the censor's optimal Stackelberg budget split allocates f* = 0.97 of its discovery budget to the domain channel once the defender runs n ≥ 2 endpoints (§8.10, Figure 7). GFWatch measurements show hundreds of thousands of domains persistently blocked.

2026-maiti-block-a-mole-sustainability-frontier §1, §3, §6, §8.10, Figure 7 detection

Simulator validation shows a sharp phase transition in (α,T)-availability at the sustainability frontier β* ≤ 1. For a canonical configuration of n=8 endpoints and buffer kmax=8, the interval frontier is β* ≈ 0.65, while the time-average crossing is near β ≈ 0.8. Sweeping rotation speed µ/λa from 1 to >100 does not move the frontier — the availability contour over the (β, µ/λa) plane is entirely vertical (Figure 3). The closed-form availability law (Theorem 1) matches simulation to three significant figures across all tested β (Table 2). A rotation-only defender at beta = 1.5 stays below 0.25 availability even at the largest buffer and rotation speed tested, while a domain-aware defender at beta = 0.6 with the same n, mu and pool exceeds 0.95.

2026-maiti-block-a-mole-sustainability-frontier §8.2, §8.5, Figure 2, Figure 3, Table 2 evaluation

Ephemeral defenses were integrated with a WireGuard fork and deployed as Mullvad VPN's 'DAITA' (Defense Against AI-guided Traffic Analysis) opt-in feature across Android, iOS, macOS, Linux, and Windows for over one year, serving a growing number of thousands of daily users. Individual defenses are derived deterministically from seeds in 43.6 ± 4.7 ms on a commodity laptop, making per-connection unique defenses practical at VPN scale.

2026-pulls-ephemeral-network-layer-fingerprinting §1, §7 deployment

Ephemeral blocking defenses reduce DF accuracy from 89.0% (undefended) to 10.2% and RF from 90.1% to 14.7% with standard 30-epoch training, at 97.5% bandwidth and 68.4% delay overhead; under infinite training, DF rises to only 29.2% and RF to 24.3%, still far below undefended baselines of 92.7% and 94.7%. Defenses are tunable at deployment time by adjusting Maybenot framework-wide limits, enabling overhead-vs-protection trade-offs without redeployment.

2026-pulls-ephemeral-network-layer-fingerprinting §5.2, §5.3, Table 1, Table 2 defense

The ephemeral property — using a unique seed-derived defense per connection — prevents attackers from training classifiers on the exact deployed defense variant. Stacked combinations with height H=5 from N=1,000 base defenses yield 6.88×10^25 unique defenses (polynomial growth O(N^{2H})). Attacks trained on ephemeral defenses also generalize significantly better across other randomized defense families than attacks trained on static defenses.

2026-pulls-ephemeral-network-layer-fingerprinting §3.3, §5.4 defense

With infinite training time, Laserbeak achieves 93.5%, 95.9%, and 95.9% accuracy against ephemeral padding, FRONT, and Interspace respectively, compared to 96.5% undefended — confirming that padding-only defenses provide no meaningful protection against a sufficiently trained deep-learning WF adversary. Only ephemeral blocking defenses retain measurable protection, reducing Laserbeak to 71.8% accuracy under infinite training versus 96.5% undefended.

2026-pulls-ephemeral-network-layer-fingerprinting §5.3, Table 2 evaluation

MIRAGE's differentially private routing function provably bounds adversary inference: for a routing protocol satisfying ε-DP with ε = ln(4), any hypothesis test achieving a true positive rate of 80% necessarily incurs a false positive rate of at least 20%. The TPR-to-FPR ratio is bounded by e^ε for any ε-DP routing function, providing a formal privacy guarantee against routing-level statistical disclosure attacks.

2026-ratliff-mirage §V–VI defense

Embedding explicit TTL values in mesh-routed messages leaks proximity information — a recipient can infer that a high-TTL message originator was recently nearby. MIRAGE mitigates this with memoryless TTLs: carriers independently discard messages with probability q per epoch, implementing a branching process with replication factor R ≤ nmax·(1−q). Setting q > 1 − 1/nmax ensures sub-critical message extinction with expected lifetime ≈ −ln(nmax)/ln(R) epochs.

2026-ratliff-mirage §VI-C defense

MIRAGE delivers 15× more messages than random-walk protocols and significantly outperforms probabilistic flooding in delivery rate. On the pedestrian YJMob100K dataset at p=0.6, MIRAGE achieved a delivery rate of 36.9%, compared to 9.1% for probabilistic flooding (4.1×) and 3.2% for handoff (11.5×). MIRAGE incurs substantially lower network load than maximal flooding (86.9% delivery) while maintaining better delivery rates than all non-flooding baselines.

2026-ratliff-mirage §VII-B evaluation

MIRAGE constructs a global mobility graph using locally differentially private per-user submissions, requiring only O(ln(|M|/β) / (α²ε²)) users to achieve per-edge accuracy α with probability 1−β. For a 100-district map with ε=0.05 and α=0.5, fewer than 1 million users suffice for top-2 district reporting; for top-3 districts the requirement drops to under 200K users.

2026-ratliff-mirage §VI-B defense

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

Scylla's AONT-based fragmentation closes the C3 condition for Fragging (fragment-drop as a tagging mechanism) but does not close the retransmission timing side-channel in Nym: per-fragment acknowledgments still allow a malicious recipient to measure fragment-set completion time and distinguish dropped-fragment senders. Full mitigation requires replacing per-fragment acknowledgments with per-message acknowledgments that retransmit all fragments under a fresh identifier only after the complete message is confirmed received.

2026-schadt-breaking-partially-fixing §5.5 defense

Dodge-mimic reduces vRF deep-learning classifier accuracy to 10.6–15.6% for k=10 anonymity sets, matching the theoretical k-anonymity bound of (100/k)%=10%. The heuristic Walls Have Ears attack also fails: WHE-seg peaks at ≤8.5% and WHE-cyk at 9.7%, confirming that equalized cycle sizes leak no per-video information.

2026-witwer-dodge-client-side-framework §5.2, Table 2 defense

In live deployment with k=10 anonymity sets, video+audio+ABR, Dodge-mimic causes zero playback stalls in approximately 68% of sessions; for the remaining 32%, the mean initial stall time is 1.6 seconds — an extension of normal startup delay rather than mid-playback interruption — and all cycles complete roughly 20 seconds before playback ends even with large trailing cycles.

2026-witwer-dodge-client-side-framework §5.3, §5.3.3 evaluation

Small random per-session perturbations (±bytes per range) to Dodge-mimic extended manifests fully compensate for HPACK header compression in HTTP/2, yielding vRF accuracy of 10.5–14.2% for k=10 — indistinguishable from HTTP/1.1 results and near the theoretical bound. Because HTTP/3-QUIC uses analogous QPACK compression, the approach generalizes to the dominant future transport.

2026-witwer-dodge-client-side-framework §5.4, Table 3 defense

With 32 worker threads, Dodge-mimic anonymity-set bandwidth overhead (as percentage of undefended video content) ranges from 21.0%±0.8 at k=2 to 73.1%±2.8 at k=10 and 107.4%±1.5 at k=20; beyond k=10–16 threads, additional parallelism yields diminishing returns and can increase overhead due to thread-boundary fragmentation of the optimal search space.

2026-witwer-dodge-client-side-framework §5.1, Table 1 evaluation

The hash-based filtering strategy shifts covert channel security from algorithm secrecy to key secrecy per Kerckhoffs's principle: even when the covert algorithm is fully exposed, an adversary without the pre-shared key cannot identify which sparse subset (~1/2^L) of packets carries covert data, transforming the detection problem from statistical pattern recognition to exhaustive key-space search. The authors recommend a minimum Input Key length of 128 bits.

2026-zou-hiding-trees-forest §3.2, §4.1, §4.2.4 defense

Fragmenting large server responses across multiple independent TCP connections each below the ≈15–20 KB threshold circumvents the freeze, but at severe cost: downloading a 50 MB file requires approximately 2,560 separate TCP connections, which is operationally suspicious and significantly degrades throughput.

2025-hyperion-cs-censor-has-new Notes defense

In a 600-node simulation on a 25×25 grid representing a city-wide blackout environment, Anix messages reached over 90% of users within 23 simulation steps (~23 hours) even when adversarial Sybil nodes composing 2% of the network refused to forward messages authored by legitimate users. The simulation modeled a 5-day blackout with 120 one-hour steps.

2025-kamali-anix Abstract; §5.2; §6 evaluation

Anix provides two cryptographically distinct identity revocation primitives: soft revocation rotates a user's identity key pair and re-notifies only the retained subset of trusted contacts via encrypted unicast, silently excluding the revoked party; hard revocation broadcasts a signed certificate containing the compromised public key components, instructing all contacts to reject both the revoked identity and any downstream identities produced through subsequent soft revocations.

2025-kamali-anix §4.3 defense

Standard ECDSA signature schemes are vulnerable to public key recovery attacks that allow an adversary to recover the signer's public verification key from any signature, linking all pseudonymous messages authored under different one-time pseudonyms back to a single user identity. This attack succeeds without any side-channel — it operates solely on the message and its ECDSA signature.

2025-kamali-anix §4.2 detection

Rangzen's transitive trust scheme suffers from two structural defects: diminishing trust (each relay hop multiplicatively reduces a message's trust score, degrading trustworthy messages from distant authors) and path dependency (the same message accrues different trust scores depending on which route it traveled, making scores incomparable across recipients). These defects prevent any user from gauging network-wide endorsement of a message.

2025-kamali-anix §2.3 evaluation

Lantern's proxyless protocol accounted for approximately 40% of its traffic during the June 2025 Iran shutdown, demonstrating that a direct-server / proxyless transport mode provided a significant load-bearing fallback when conventional proxy infrastructure was blocked by centralized DPI enforcement.

2025-miaan-stealth-blackout Executive Summary — Resilient Response evaluation

The framework confines active traffic shaping to the first N seconds of a connection (N is a user-defined parameter, e.g., N=10), after which normal unmodified traffic resumes. The authors hypothesize that this design keeps per-session throughput and latency overhead negligible, since the shaping window is a small fraction of total connection time; N can be extended to the full session if the censor is believed capable of classifying beyond early traffic.

2025-pereira-extended §2.3, §3 defense

The proposed framework operates as a transparent shim between application and network layers, enforcing a configurable schedule over packet size, timing, and burst patterns. The shaping logic is transport-agnostic — applicable across TCP, UDP, QUIC, and TLS — and activates only after the underlying protocol handshake completes, making it reusable across heterogeneous circumvention stacks.

2025-pereira-extended §2.1 defense

The framework is designed for adoption into existing censorship-resistant systems in the same manner as uTLS — as a drop-in Go library requiring minimal code changes. Primary integration targets are Tor pluggable transports and WireGuard-based VPNs that currently lack built-in traffic obfuscation. Predefined hand-crafted schedules are provided alongside GAN-generated ones to enable developer stress-testing without model inference.

2025-pereira-extended §2.3 defense

Security arguments for existing circumvention systems are based on ad-hoc adversary models that are often incomplete or unrepresentative of real-world adversaries, leading to allegedly secure designs that fail against relatively straightforward attacks. Protocols that substitute or parasitize a cover application's encrypted traffic channel fail against application-aware adversaries who observe or induce violations of application-specific behavioral invariants — a weakness that pre-trained classifiers on custom traces fail to surface.

2025-pereira-position §1, §2 defense

An adversary's false positive rate against a circumvention tool depends critically on the statistical properties of background traffic; if background traffic is modeled inaccurately (e.g., with toy uniform distributions), formal detection bounds are not meaningful. The paper proposes a hybrid pipeline: train NetDiffusion on real packet-level traces from campus networks or backbone providers, sample synthetic background traffic, extract empirical mean/variance, and integrate those distributions into EasyCrypt formal models to produce statistically grounded detectability proofs.

2025-pereira-position §3 evaluation

The paper proposes modeling HCS undetectability as a simulation-based cryptographic distinguishability problem: if traces produced by the real-world HCS channel are computationally indistinguishable from ideal-world application-channel traces (T_HCS ∼ T_simulator), the HCS achieves provable security against any adversary — passive or active. The simulation paradigm is parametric in adversary capability, meaning a single proof covers the full spectrum from passive SNI monitoring to active DPI.

2025-pereira-position §1, §2 defense

The blocking-resistance of CenPush derives from the collateral damage a censor would incur by blocking APNs or FCM: doing so would break push notifications for every app on iOS or Android respectively. This is the same collateral-damage deterrent mechanism that makes CDN-based domain fronting and TLS-over-CDN transports resilient, applied to the control plane rather than the data plane.

2025-sharma-cenpush §2, §4 defense

CenPush uses mobile platform push-notification services (APNs, FCM) as a blocking-resistant control channel for distributing fresh proxy IPs and client configuration to users in censored regions. Push notification infrastructure is already widely deployed, has high collateral-damage cost to block, and is a server-push channel — meaning the client never has to initiate a query to an out-of-band endpoint that a censor could block.

2025-sharma-cenpush §1, §3 defense

CenPush is implemented and evaluated specifically for Tor bridge distribution, replacing the existing polled bridge-line fetching with push delivery. The design is presented as a general mechanism applicable to any circumvention tool that needs to push fresh proxy addresses to clients — not just Tor bridges — whenever censors block the tool's normal update channel.

2025-sharma-cenpush §3, §5 deployment

The paper defines Unauthenticated Push (UP) channels as a distinct archetype from signaling/rendezvous channels, characterized by three properties: strictly unidirectional delivery, no client authentication or account association required, and higher bandwidth (kilobytes to megabytes) to support software updates rather than just minimal proxy-address exchanges. This design deliberately shifts operational-security burden onto senders to approach receiver anonymity.

2025-vines-extended §1–§3 defense

Among surveyed channels, Skyhook, PushRSS, SQS, AMPCache, and Meek satisfy all three UP channel properties (unidirectional, no client auth, higher bandwidth); CloudTransport and Raven do not because they require authenticated user accounts; Tor's email- and Telegram-based bridge distribution also fails the no-auth requirement. The analysis was prompted in part by the 2022 GFW entropy-based blocking event, which required software updates to be pushed to users before fully-encrypted protocols could resume functioning.

2025-vines-extended §3 defense

Three open-source DPI tools (Zeek, libprotoident, nDPI) fail to identify 93–100% of UPGen flows across all tools. libprotoident misidentified 7% of UPGen flows as RTMP; nDPI and Zeek produced zero false labels. On a real-world MAWI/WIDE backbone capture, Zeek failed to recognize 90% of flows and nDPI failed on 67%, confirming that unidentified-protocol traffic is common in the wild; allowlisting without significant collateral damage (≥4%) is infeasible.

2025-wails-censorship §4.4, Table 5, Table 6 evaluation

UPGen's generator samples 18 independent parameters to produce 4.2×10^22 distinct structured encrypted protocols (entropy 38.4 bits). Each proxy is assigned a unique generated protocol, so identifying one protocol exposes only a single proxy. The generator was designed by studying 27 real-world encrypted protocols and sampling from observed structural patterns (greeting strings, handshake patterns, field orderings, key encodings).

2025-wails-censorship §2.4, Table 1 defense

Custom CCAs that deviate from standard TCP/QUIC congestion response fundamentally contradict the core circumvention principle of traffic indistinguishability: by failing to back off under congestion signals, they produce traffic patterns that diverge from the vast majority of Internet flows that censors value, eliminating the collateral-damage protection that makes circumvention tools hard to block wholesale.

2025-wang-custom §1, §6 detection

Shaperd's adaptive blocking-detection mode can integrate with external blockage-detection tools (e.g., Troll Patrol) to detect when a constraint set is no longer effective and automatically switch to an alternate constraint set, changing packet patterns to restore connectivity without user intervention.

2025-wilson-extended Appendix A defense

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

The paper evaluates two short-term mitigations—TCP delayed ACK on the proxy server and connection multiplexing—but finds both are limited: delayed ACK produces atypical ACK timing that may itself be fingerprintable, and multiplexing only adds entropy without eliminating the RTTdiff signal. Critically, obfs4 and ScrambleSuit's delay-based timing obfuscation are described as 'fundamentally limited' because they manipulate inter-arrival times without eliminating the underlying transport/application-layer session misalignment. The paper concludes no existing obfuscation scheme provides a principled defense against timing-based proxy fingerprinting.

2025-xue-discriminative §VII, §I defense

The authors propose a 'shim' pluggable transport that splits client traffic across N PT connections using unmodified existing PT bridges as proxies and a gateway bridge that correlates streams back into a Tor circuit via the Turbo Tunnel reliability pattern. This architecture enables all existing and future PTs to benefit from traffic splitting without modifying each PT's client or server code individually.

2024-lorimer-extended §3 defense

Web security vulnerabilities whose exploitation depends on parser divergence between two co-located systems are structurally isomorphic to censorship circumvention attacks, where the censor acts as the frontend parser and the destination server as the backend. The authors demonstrated this by directly converting all HRS test vectors from prior security research into circumvention probes with no modification, showing that censorship-circumvention techniques can be systematically constructed from existing vulnerability corpora.

2024-m-ller-turning §3, §8 (Conclusions) defense

The TLS-Attacker suite is being extended to cover QUIC and DTLS 1.3 under a universal analysis framework that reuses existing Workflow Trace and Modifiable Variable machinery with only protocol-specific components added. As of 2024 the QUIC dialect is functional, making TLS-Attacker the only open-source tool that can fuzz TLS, DTLS, and QUIC handshakes under a single scriptable API.

2024-niere-tls-attacker §2 Development / §3 Outlook evaluation

TLS-Attacker's Workflow Traces and Modifiable Variables mechanisms allow testers to specify arbitrary protocol flows and apply field-level modifications — including adding, removing, or overwriting individual TLS message fields — without breaking the internal TLS state machine. This makes it the standard instrument for probing how DPI systems and active-probing detectors respond to non-standard or mutated TLS handshakes.

2024-niere-tls-attacker §1 Introduction evaluation

TLS-Scanner, a subproject of the TLS-Attacker suite, automates handshake probes across deployed TLS hosts and has been used in published IPv4-wide scanning studies. It surfaces supported protocol versions, enabled extensions, and known vulnerabilities, providing a ready-made audit tool for circumvention infrastructure operators.

2024-niere-tls-attacker §2 Impact (Academic Perspective) evaluation

Separating the Broker role (a server that holds and manages bridge information) from both the rendezvous channel and the censorship evasion system enables modular protocol design: the rendezvous carrier can be swapped independently of the proxy system. The authors identify broker authentication and multi-broker load distribution as open problems not addressed in the current prototype.

2024-vilalonga-looking §3.1, §5 defense

The paper identifies that circumvention systems relying on long-lived, consistent proxy servers are fundamentally vulnerable to host-based temporal detection regardless of per-flow obfuscation quality, and recommends adversarial examples, ephemeral obfuscation servers, and programmable or polymorphic protocols as countermeasures. Snowflake's volunteer-browser proxy architecture—where proxies are ephemeral and addresses are not reused—is highlighted as inherently more resistant to host-based classification than static bridge designs like obfs4.

2024-wails-precisely §I, §VI discussion, §VII defense

DeResistor's two-objective fitness function (balancing evasion success and detection probability) reduces flow-level detection rates from 96.27% → 45.06% against China's GFW, 99.50% → 34.93% against India, and 99.50% → 49.22% against Kazakhstan over 5 training generations, while in all cases preventing TRW from reaching an IP-block decision that would terminate training.

2023-amich-deresistor §5.1, §6.2, Table 1 defense

Interleaving a single normal benign flow (jump size J=1) after each detected probe prevents the TRW likelihood ratio from converging to the IP-block threshold across all 11 simulated censors and all three real-world censors tested; setting J>1 risks triggering a history-aware TRW reset that can paradoxically accelerate IP-level detection.

2023-amich-deresistor §5.2, §6.2 defense

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

DeTorOS's security relies on the honest-but-curious model: if the onion service refuses to participate or lies about its circuit, the client receives no avoidance guarantee. The paper explicitly flags this as an open limitation and notes it cannot be closed without either requiring a TEE on the onion service side or fundamental protocol changes.

2023-arora-detor-onion §6 defense

The paper argues that an effective counter to translation censorship is to actively trigger the Streisand effect: publishing detected censored content side-by-side with the original on a public website causes the censored text to reach a broader audience — including people who would not have read the censored version — and makes the censorship itself backfire. Censors deliberately avoid publicizing removals precisely to prevent this outcome.

2023-streisand-where §1 Introduction / §5 Conclusion defense

The proposed crowdsourced system runs multiple isolated Geneva training pools on a controlled server — one pool per censorship system (initially China and Iran) — and instructs volunteer browsers via JavaScript to send forbidden requests to isolated ports, with no download or software installation required from the user. The server monitors per-strategy success or failure to drive genetic evolution entirely from the server side.

2023-tran-crowdsourcing §3 Design defense

The system is designed to protect crowdsourced volunteer privacy by storing only AS-level granularity alongside randomized short-lived client identifiers, explicitly discarding source IP addresses and any browser-identifying information. AS-level resolution is sufficient for server-side evasion because strategies are evolved per-censor-ASN rather than per-user.

2023-tran-crowdsourcing §3 Design — Protecting Users defense

Server-side censorship evasion strategies require zero client-side changes: clients bypass censorship without installing software or even being aware of the evasion, and this approach has been adopted in production tools including Psiphon's packetman. The packet manipulations exploit weaknesses in how censors track or tear down TCP connections, occurring entirely at the server during the three-way handshake.

2023-tran-crowdsourcing §1 Introduction defense

In a Rust implementation evaluated over 10,000 runs with 3,600 bridges (1,800 open-entry single-bridge buckets and 600 hot-spare three-bridge buckets), Lox's trust promotion protocol incurs the highest latency at 364.2 ms response time and 378 kB response size due to the encrypted migration hashtable, but this operation occurs only once per user. All other protocols complete in under 16 ms with request sizes under 3.4 kB.

2023-tulloch-lox Table 4, §5.2 evaluation

Lox's trust level scheme (L=0 through L=4, requiring 30, 14, 28, 56, and 84 days respectively per level before upgrading, per Table 2) with blockage inheritance — invited users inherit their inviter's blockage count d — prevents a censor from resetting their reputation through self-invitation after causing blocking events, while users with d ≥ 4 become ineligible to migrate, capping the damage a persistent infiltrator can do.

2023-tulloch-lox §4.2.1, §2.4.2, Table 2 defense

Lox uses Chase et al.'s keyed-verification algebraic MAC anonymous credentials in a single-issuer/verifier setting with jointly-chosen credential IDs (neither party can unilaterally select them), so a fully compromised Lox Authority cannot link credential showings to specific users or reconstruct the social graph — the LA learns only that a shown credential was authentically issued.

2023-tulloch-lox §3.1, §4.2 defense

Proteus does not yet implement normalized or randomized error responses, and the authors explicitly flag this as a known gap: without configurable error handling, the protocol may be identifiable by an active prober who can distinguish the proxy's error behavior from that of the legitimate service being mimicked.

2023-wails-proteus §5 defense

Marionette, the prior programmable protocol system, executes user-specified plugin code in a generic Python runtime, making proxies and clients vulnerable to a malicious or buggy protocol distributor and creating a single point of failure in distributed networks like Tor. Marionette also lacks support for multiple simultaneous protocols and version upgrades, limiting its ability to respond to changing censorship rules across heterogeneous client populations.

2023-wails-proteus §4 evaluation

A complete Noise NK handshake protocol — including Curve25519 ECDH key exchange with server authentication, HMAC-based key chaining, and ChaCha20-Poly1305 AEAD-encrypted data phase — was expressible in Proteus in less than 4 hours, demonstrating that a safety-bounded DSL with built-in crypto primitives and declarative message-format definitions is sufficient to prototype complex cryptographic transport protocols rapidly.

2023-wails-proteus §3.3 evaluation

Proteus supports simultaneous execution of multiple Protocol Specification Files (PSFs) on a single server, selecting the correct protocol version by running all candidates in parallel on a shared read buffer and eliminating candidates as they fail to parse client messages. This enables servers to support legacy clients while deploying new evasion protocols, and to serve clients in different censorship regimes with localized protocol variants, without requiring synchronous client/server upgrades.

2023-wails-proteus §2.4 defense

Meteor encodes bits by embedding a PRG-masked random value into the token-sampling randomness of a generative model, recovering bits proportional to the shared prefix length of the sampled interval. Expected throughput per sampling event is asymptotically within 1/2 of the Shannon entropy of the channel (proven in Appendix A), so Meteor automatically adapts to high entropy variability without explicit signaling or padding.

2021-kaptchuk-meteor §5 Meteor / §5.2 defense

Meteor is proven secure against chosen-hiddentext attacks: any PPT adversary distinguishing Meteor output from honest model output can be reduced to breaking the underlying PRG. The scheme produces stegotext provably indistinguishable from the generative model's own output distribution, and requires only a shared public model — not a secret channel — making the model analogous to a common random string. On GPU the encoding overhead is ~1× model-load time; on CPU ~4.6×; on mobile ~49.5×.

2021-kaptchuk-meteor §5.2 / §6 / Table 2 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

The paper presents 11 purely server-side censorship evasion strategies requiring zero client-side software, successfully bypassing censorship in China, India, Iran, and Kazakhstan across DNS-over-TCP, FTP, HTTP, HTTPS, and SMTP. All strategies manipulate only TCP handshake packets (primarily the SYN+ACK) and were verified against 17 versions of 6 client operating systems (Windows XP–Server 2018, MacOS, iOS, Android, Ubuntu, CentOS) with unmodified clients.

2020-bock-come §1, §5, Table 2 defense

The dnstt DNS-over-HTTPS tunnel, built on a KCP Turbo Tunnel session layer, achieved download speeds of 130 KB/s using Google and Cloudflare DoH resolvers and 30 KB/s using Quad9, compared to iodine's maximum of 2 KB/s over the same operators' UDP DNS resolvers — a 15–65× improvement. DNS-over-HTTPS hides message contents from the censor, removing the two main classical DNS tunnel detection vectors: unusual DNS message structure and plaintext tunnel domain names in queries.

2020-fifield-turbo §3.4 evaluation

Simultaneous upload and download of a 10 MB file took 10.6 s over TCP-encapsulated QUIC, 23.3 s over traditional meek, and 34.9 s over meek with encapsulated QUIC (Table 1), showing that naively adding a QUIC session layer to meek degraded throughput by approximately 50% relative to unmodified meek. Performance was sensitive to HTTP body size limits and request-thread count, but the root cause remained uncertain.

2020-fifield-turbo Table 1, §3.2 evaluation

Geddes et al. demonstrated that acknowledgement packets in covert-channel circumvention systems can be identified through timing characteristics and selectively interfered with to disrupt the tunnel [§4.3, CCS 2013]. A Turbo Tunnel session layer adds fixed-overhead headers and periodic ACK/keepalive traffic that may produce distinctive timing patterns absent in legitimate flows, potentially increasing susceptibility to traffic-shape classifiers.

2020-fifield-turbo §2 detection

Turbo Tunnel inserts an interior session/reliability protocol (KCP or QUIC) between the obfuscation layer and user streams, decoupling end-to-end session state from any single transport connection. A session survives TCP termination, proxy rotation, or unreliable carriers by retransmitting lost packets over a new connection bearing the same session identifier. The pattern was implemented in obfs4, meek, and Snowflake, with Turbo Tunnel–enabled Snowflake shipping in Tor Browser alpha releases 9.5a13 (desktop) and 10.0a1 (Android).

2020-fifield-turbo §1–§2 defense

MIMIQ requires modifications only to a single trusted network (an ISP or enterprise AS): an address allocation server and several critical edge switches. Remote QUIC servers and the wider Internet require no changes. The authors argue ISPs have financial incentives to deploy MIMIQ as a privacy-enhancing service, and that QUIC's rapid adoption (600K+ QUIC-enabled domains, 1.6M QUIC-capable IPs as of 2020) means server-side support is increasingly given.

2020-govil-mimiq §5 deployment

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

SiegeBreaker's session bootstrapping (from initial email to installed SP redirection rule) averaged 3–4 seconds across 100 trials, with the dominant delay attributed to email handling (SMTP connection, Selenium composition) rather than network latency; this setup cost is not included in the download-time benchmarks. The auxiliary ping-based switch-selection signal encodes 48 bits across three ICMP header fields (IP-ID, ping sequence number, ping identifier), requiring ~281 trillion spoofed ping packets per client–OD pair to brute-force.

2020-sharma-siegebreaker §4.4, §5.2 evaluation

SiegeBreaker explicitly acknowledges two unresolved attack vectors: (1) latency-based traffic analysis attacks (forced-asymmetry / RAD-style), which the system does not mitigate, and (2) website fingerprinting attacks against the proxied traffic, for which no defense is implemented. Additionally, the email-based control channel is vulnerable to a censor who can delay or block emails to the controller's address, disrupting rule installation before the client's SYN packet arrives.

2020-sharma-siegebreaker §3, §4.1, §4.3 detection

Prior decoy routing deployments suffered severe throughput degradation: the TapDance ISP pilot reported average client throughput of only ~5 KB/s, making it unsuitable for most web content; other DR prototypes restricted evaluation to files under 1 MB in controlled lab settings, with some reporting over 30 seconds to load home pages under 1.5 MB in size.

2020-sharma-siegebreaker §3, §2.3 evaluation

All prior decoy routing systems (Cirripede, Telex, TapDance, Slitheen, Waterfall) require the DR to inspect every traversing flow — either all TCP SYN packets or all TLS flows — to identify DR requests, creating a privacy breach for non-DR users and a computational bottleneck. SiegeBreaker eliminates this by using an out-of-band email pre-registration (encrypted to the controller's 2048-bit RSA public key) that pins the controller's inspection rule to a single client-IP/OD-IP/ISN triple, so only authenticated potential DR flows are ever redirected.

2020-sharma-siegebreaker §3, §4.3 defense

SiegeBreaker achieves near-native TCP performance in Internet experiments: average download time for Alexa top-500 home pages via SB was 1.8 s versus 1.7 s for direct wget, across 500 concurrent client instances; bulk downloads of 1 GB files over a shared 1 Gbps link showed SB and native TCP sharing bandwidth almost equally, and throughput remained stable under 15 Gbps of cross-traffic or 50,000 parallel flows on the SDN switch.

2020-sharma-siegebreaker §5.1, §5.2 defense

SymTCP uses selective symbolic execution over Linux's TCP implementation (S2E + KLEE) to enumerate all packet sequences reaching 47 binary-level accept or drop points from LISTEN to ESTABLISHED, then conducts differential testing against a blackbox DPI to confirm discrepancies; the open-sourced system requires no DPI source access and covers 37 of 47 drop points within the operationally relevant handshake window.

2020-wang-symtcp §IV, §VIII.A–B defense

SymTCP generated 56,787 candidate insertion/evasion packets in approximately one hour using concolic execution over Linux's TCP stack. Evaluating a sampled set of 10,000 test cases against real DPI systems yielded 6,082 evasions against Zeek, 652 against Snort, and 4,587 against the Great Firewall of China — discovering 14 novel evasion strategies beyond those found by prior manual approaches.

2020-wang-symtcp §VIII.C evaluation

A proxy assignment algorithm derived from the Gale-Shapley college admissions game, using multi-feature utility functions across five client metrics (proxy utilization capped at T, new-proxy request rate, blocked-proxy usage, known-blocked count, client distance) achieves superior connected-client ratios and lower wait times compared to state-of-the-art rBridge in all tested ecosystem configurations (Static, Slow, Alive, Popular), without requiring knowledge of individual client types at assignment time.

2019-nasr-enemy §IV, §VII defense

A game-theoretic optimal censorship strategy — in which coordinated agents maximize a joint utility combining proxy discovery and blocking impact (equation 3, parameterized by ω) — is significantly stronger than both aggressive (immediate block) and conservative (timed-delay) heuristic strategies evaluated in prior work including rBridge; changing ω (surveillance vs. blocking preference) further modulates the damage a censor can inflict on any given distribution profile.

2019-nasr-enemy §VII-C evaluation

For 1 MB files, even at a database of only 50,000 entries, PIR responses reach 73.1 MB per retrieval, making proof-of-censorship impractical for image or video streaming content providers. By contrast, for 256-byte (Twitter-like) messages the system remains workable at 10 million files with 8.0 MB queries and 2.0 MB replies, and stays roughly constant in reply size (2.0 MB) between 500k and 10 million files.

2018-martiny-proof-of-censorship §5, Tables 2–3 evaluation

The proof-of-censorship scheme uses single-server computational PIR with homomorphic encryption so that the server, having signed both the PIR query hash and its reply, cannot selectively omit responses for a targeted file without returning garbage data. A client detecting the mismatch publishes the upload ticket, signed reply, and query seed as a compact, transferable cryptographic proof of censorship verifiable by any third party holding the server's long-term public key.

2018-martiny-proof-of-censorship §4.1 defense

On a quad-core Intel Core i5 (3.30 GHz) against a database of 1 million 256-byte messages, the prototype produces a 3.8 MB PIR query (28 ms client-side generation) and a 2.0 MB proof requiring 2.8 s of server-side processing; third-party proof validation takes 52 ms, and the 120-byte upload ticket validates in 381 µs. All client-side operations are fast enough for smartphone or JavaScript implementations.

2018-martiny-proof-of-censorship §5, Table 1 evaluation

A censoring server cannot selectively withhold PIR responses for a targeted file while honestly answering others: if a PPT algorithm A could distinguish targeted-file queries from all other queries, it would directly violate the query privacy of the underlying PIR scheme. The server's only compliant evasion strategy is an indiscriminate shutdown — refusing all queries or all signatures — which is behaviorally distinguishable and does not produce a plausible-deniability defense.

2018-martiny-proof-of-censorship §4.3 defense

Proofs of censorship are transferable and persistent: even if a content provider restores a censored file, previously generated proofs remain cryptographically valid and can serve as a reputation mechanism, a trigger for smart-contract financial penalties (e.g., Ethereum bonds), or mandatory disclosures to transparency databases such as Lumen, enabling accountability for transient or temporally-selective censorship that current transparency reports cannot capture.

2018-martiny-proof-of-censorship §6.2 policy

The Lavinia audit protocol is designed so that auditors are cryptographically indistinguishable from ordinary readers: an auditor cannot reveal her status to a server without forfeiting her own payment, and servers are therefore forced to serve content in response to every request. Any reader may additionally claim to be an auditor, and servers cannot verify such claims, further preventing selective serving.

2017-bocovich-lavinia §4.1, §4.4, §6 defense

The burn contract mechanism defends against deliberate auditor-chain termination attacks, in which a malicious actor poses as an auditor and refuses to post her secret, preventing all subsequent auditors from performing their audits. If the previous auditor fails, the current auditor can burn both her predecessor's payment and her own, receive a small fraction of those funds as incentive, and forward the chain secret to the next auditor — preventing a single compromised link from collapsing the entire revenue stream for a document.

2017-bocovich-lavinia §4.3 defense

Lavinia requires its underlying payment system to satisfy four properties for suitability in censorship-resistant contexts: (1) coercion-resistance through geo-political distribution or anonymization, (2) redeemable with a distributable secret, (3) time-locked escrow preventing early redemption, and (4) an append-only public log. The paper demonstrates that Bitcoin satisfies all four properties, with Zerocash extensions providing payment anonymization to prevent linking payments to specific documents.

2017-bocovich-lavinia §3.2 defense

Theorem 1 proves a dominant strategy Nash equilibrium in which all rational servers honestly store and serve all files, subject to the constraint that per-server audit payment exceeds routing cost and file-serving payment exceeds storage cost. At 2017 prices, storage hardware cost approximately $0.03/GB and bandwidth cost approximately $0.03/GB, so the minimum per-file hosting payment must exceed (η + BR) × $0.03/GB × |f|.

2017-bocovich-lavinia §5.1, Theorem 1 evaluation

Lavinia allows a publisher to publish content, submit payments, and then cease all interaction with the system — continued document availability is not contingent on the original publisher remaining online or reachable. This specifically protects against out-of-band coercion tactics such as rubber-hose cryptanalysis in the case that the publisher is captured or prosecuted.

2017-bocovich-lavinia §4.1 defense

A CAPTCHA-gated registration scheme with sequences of reCAPTCHAs at random intervals and short solve windows limits automated censor deployment. With 5 minutes spent per registration, a human adversary working non-stop for 24 hours can create at most 288 censors; combined with a 12-hour registration reset cycle, this bounds the adversary's censor accumulation rate.

2017-heydari-scalable §IV-B defense

For complete blockage (>99%) over 10 hours, the adversary requires a swarming ratio of 12.8, translating to 128,000 censors against a single server with 10,000 CoAs. Scaling to a 10-server, 10-interface deployment forces the adversary to operate 106,700 humans in parallel; with a 5-minute CAPTCHA registration and a 12-hour reset cycle, achieving complete blockage within 10 hours requires 1,067 non-stop human operators in the first two hours.

2017-heydari-scalable §IV-A, §IV-B evaluation

A credit-based accounting method dynamically assigns users to larger groups as their trust score accumulates (credit increases by G−1 per unblocked interval), requiring a user's credit to be twice the group's risk before joining. This reduces the total number of CoAs needed while making it costly for censor agents to infiltrate large groups, since they must wait through many clean intervals before the group reaches exploitable size.

2017-heydari-scalable §IV-D defense

A proof-of-concept Linux prototype using UMIP (open-source MIPv6) with three routers and five commodity machines (2.4GHz Intel Core 2 Duo, 4GB RAM) demonstrated correct CoA rotation every 10 seconds. Signaling overhead was reduced to one-third of standard MIPv6 by eliminating return routability messages; per-packet transmission overhead was 24 bytes (IPsec ESP), identical to the baseline secure-channel cost, yielding zero net overhead attributable to the MTD mechanism.

2017-heydari-scalable §IV-E evaluation

The MI-MTD framework uses Mobile IPv6 Care-of Addresses (CoAs) rotated among randomized user groups every shuffling interval. With 1,000,000 users, 5,000 censors, and 10,000 CoAs (swarming ratio φ=0.5), per-interval access probability is 60.88%; over one minute with 10-second shuffling intervals, blocking probability drops to approximately 0.358%, meaning users retain ~99.6% chance of access.

2017-heydari-scalable §IV-A defense

ADVENTION's split-path design — fetching publisher content via relay and ad requests via the direct path — raises average ad-set overlap from 28% (Tor) to 70%; combining ADVENTION with Intelligent Relay Selection (language-matched relay) further increases average overlap to ~80%. For blocked sites, ADVENTION with IRS raised ad relevance from ~16% to 100%.

2017-javaid-online §5.1, §5.3, §5.4 evaluation

ADVENTION provides up to 47% improvement in average page load time (PLT) compared to Tor, because ad requests — which are often on the critical rendering path — are served over the direct channel rather than through the relay. The exact improvement depends on webpage structure and bottleneck resources.

2017-javaid-online §5.1, Figure 4 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

TCP segment splitting and out-of-order delivery evades DPI classification in the testbed, T-Mobile, and Iran, but fails against the GFC—which performs extensive packet validation and correctly reassembles reordered streams—and AT&T, which uses a transparent HTTP proxy that normalizes all traffic before inspection. Payload splitting to one byte in the first packet is sufficient to defeat packet-count-limited classifiers.

2017-li-lib-cdot-erate §4.3, Table 3 evaluation

lib·erate's TTL-limited inert packet insertion—sending a decoy packet with TTL set to expire at the middlebox but carrying a misclassifying payload—successfully evades classification in a carrier-grade testbed DPI device, T-Mobile's Binge On, and the Great Firewall of China, but fails against Iran's censor and AT&T (Table 3). When bilateral server support is available, inserting a single dummy packet at flow start evades classification in all four deployments.

2017-li-lib-cdot-erate §4.3, Table 3 evaluation

INTANG, a measurement-driven tool that caches the best-performing TCP evasion strategy per server IP, achieves an average success rate of 98.3% (range 93.7%–100%) from vantage points inside China. Four combined new strategies — Improved TCB Teardown, Improved In-order Data Overlapping, TCB Creation + Resync/Desync, and TCB Teardown + TCB Reversal — each independently achieve average success rates of 94.5%–96.2% inside China and 84.6%–92.7% outside China, with Failure 2 rates below 1.1%.

2017-wang-your §7, Table 4 defense

A censor tracking which deleted posts are resurrected can apply Bayesian inference to identify content-preservation system users: for each resurrected post r observed by set O(r), each observer's suspicion score updates by factor (|O(r)|−1)/|O(r)|, while observers of non-resurrected deletions can be ruled out with certainty. The attack requires only that the censor join the preservation system with a few sock-puppet accounts spread across multiple followed-user lists.

2016-douglas-ghostpost §5 detection

A censor with platform-side control can definitively confirm a single suspected user by injecting a unique fake post visible only to that user, then querying the preservation system for resurrected posts attributed to that fabricated author. Presence of the fake post in the resurrection feed is binary confirmation of user membership. This targeted attack defeats automated post-alteration countermeasures when a human examines the result.

2016-douglas-ghostpost §5.1 detection

Simulation on a 1,000,000-user scale-free Weibo topology shows that at 1% GhostPost user adoption the system preserves over 70% of postviews against the daytime censor (2-hour median deletion) and nearly 90% against the nighttime censor (10-hour median deletion). Even a highly aggressive censor deleting posts within 30 minutes on average cannot prevent a 1.5% GhostPost deployment from resurrecting the majority of postviews. Steep coverage gains plateau around 0.5% adoption, after which marginal returns diminish.

2016-douglas-ghostpost §4.2, Figure 1 evaluation

GhostPost's client-server coordination channel transfers only metadata and small text payloads, making it neither bandwidth-intensive nor latency-sensitive. The paper explicitly concludes that 'practically any means of communication, including low-performance covert channels, are adequate' for the coordination channel, enabling operation over DNS tunnels, steganographic channels, or other constrained transports when the central server's HTTPS endpoint is blocked.

2016-douglas-ghostpost §2.1 defense

Adding a DPI apparatus with true positive rate TPR and false positive rate FPR creates three ordered thresholds Fam ≤ Fab ≤ Fmb governing censor strategy: allow all traffic (CTP ≤ Fam), deploy the apparatus (Fam < CTP ≤ Fmb), or block all traffic (CTP > Fmb). The apparatus does not qualitatively change the Nash equilibrium structure; it only shrinks the CTP range the circumventor can sustain, with the ordering Fmb ≥ Fab ≥ Fam holding whenever TPR ≥ FPR.

2016-elahi-framework §4.3 detection

A censor can mount a zero-collateral-damage flooding attack by injecting fake CRS-protocol-conformant traffic into open channels, inflating the apparent CTP and evicting real circumvention traffic to throttled or sacrificial protocols. If injection is costless the censor can drive real circumvention throughput to zero while keeping all channels nominally open; the attack is equally effective against both throttling and dumping CTP control strategies.

2016-elahi-framework §6.1 detection

In a single-round censorship game the only Nash equilibrium that keeps the channel open requires the circumvention traffic proportion (CTP) satisfy CTP ≤ F, where F = (βant+βbnt)/(αact+αbct+βant+βbnt). In repeated indefinite games a stable equilibrium exists at CTP = Z = (1−p)·CTPmax, where p is the per-round continuation probability, allowing a non-zero proportion of circumvention traffic to flow indefinitely without triggering shutdown.

2016-elahi-framework §4.1–4.2 defense

The optimal multi-protocol CRS traffic allocation distributes circumvention traffic across n cover protocols proportionally to each protocol's non-circumvention traffic volume (CTPi = Li · CTP/(1−CTP)), keeping every individual protocol below the blocking threshold. This makes individual protocol channels independently optimizable, with the sole selection criterion being maximizing cover traffic volume L rather than any other protocol property.

2016-elahi-framework §5.1 defense

Throttling—capping total CRS traffic at Fab and withholding surplus—strictly dominates dumping surplus traffic onto a sacrificial protocol that will subsequently be blocked. Table 2 shows that at CTP = Fab·1.05 the circumventor's relative utility drops to 0.88 of the Fab baseline when dumping, while throttling preserves all open protocols; under a censor flooding attack dumping additionally loses protocol n entirely, making throttling the dominant strategy in both attack and no-attack conditions.

2016-elahi-framework §5.1.3, Table 2 defense

Castle structurally avoids all three covert-channel pitfalls identified by Geddes et al.: architecture mismatch is avoided by supporting both client-server and P2P modes; channel mismatch is avoided because RTS games implement application-layer reliability over UDP (matching proxied TCP requirements, unlike VoIP), blocking selective-drop denial-of-service attacks; content mismatch is avoided because legitimate RTS traffic has high natural variance driven by map, strategy, and player count.

2016-hahn-games §6.3 defense

Camouflage bypassed GFW censorship in China across one month of daily testing with no plugin blocked. The GFW's primary mechanism was identified as keyword filtering on web content rather than DNS hijacking (avoided due to risk of collateral international impact). Dropbox was inaccessible inside China during testing, demonstrating that plugin substitutability is operationally necessary: at least one alternative protocol must remain reachable in any given censored environment.

2016-zarras-leveraging §5.3 evaluation

To match legitimate user behavior, the Camouflage dispatcher enforces empirically derived per-protocol session time limits: email 1–3 minutes, file sharing 5–10 minutes, instant messaging 15–20 minutes, and VoIP 20–30 minutes (Table 1). Sessions exceeding these windows produce a detectable deviation from population-level usage norms.

2016-zarras-leveraging §5.2, Table 1 defense

A single-protocol circumvention system creates a detectable anomaly: when the system is active, the traffic pattern on that protocol diverges from the same user's baseline behavior, which anomaly-based detectors can classify. Users who also legitimately use the tunneled service in daily life produce two distinct signatures — one with and one without the circumvention layer — further compounding detectability.

2016-zarras-leveraging §5.2 detection

Marionette is the first programmable obfuscation system to simultaneously satisfy all five threat-model dimensions evaluated in Figure 2: resistance to blacklist DPI, whitelist DPI, statistical-test DPI, protocol-enforcing proxy traversal, and multi-layer traffic control, while sustaining throughput above 1 Mbps (up to 6.7 Mbps). Every prior system (obfs4, ScrambleSuit, SkypeMorph, StegoTorus, FTE, JumpBox, etc.) fails at least one dimension, most commonly stateful proxy traversal or statistical-feature control.

2015-dyer-marionette §2, Figure 2 defense

On a 370-node PlanetLab deployment, Alibi Routing achieved near 100% success avoiding both the USA and China (Tables 1–2) with an average search cost of 1.0–1.66 nodes contacted (Table 4). In simulation over 20,000 globally distributed nodes, success rates were 93–100% at δ=0.5–1.0 with average search cost under 40 nodes (Table 3), capping TTL at 7.

2015-levin-alibi §6.3, Tables 1–4 evaluation

For the vast majority of source-destination pairs avoiding the USA or China on PlanetLab, Alibi Routing introduces less than 50% latency inflation; some pairs even see latency improvement due to overlay shortcutting (Figure 9). Latency inflation is relatively insensitive to the inequality factor δ when relays are successfully found.

2015-levin-alibi §6.3, Figure 9 evaluation

Property 1 proves that a peer inside a forbidden region F cannot satisfy the safety condition: appearing safe would require reporting an RTT lower than (3/c)·distance(peer,F), a physical impossibility. Property 2 follows: all trustworthy peers ignore packets routing through F regardless of attacker-controlled neighbor sets, making Alibi Routing safe without assuming honest neighbor selection.

2015-levin-alibi §5 defense

Alibi Routing fails for source-destination pairs close to or inside the forbidden region: approximately 10% of pairs cannot provably avoid China and 22% cannot avoid the USA at δ=1.0 (Figure 5), with a strong monotonic correlation between proximity to the forbidden region and the number of available relays (Figure 6). Additionally, about 50% of nodes in target regions fail the alibi condition when avoiding the USA due to its BGP routing centrality causing actual paths to transit it despite geographic distance (Figure 7a).

2015-levin-alibi §6.1–6.2, Figures 4–7 evaluation

Alibi Routing proves packets avoided a forbidden geographic region using physical impossibility: a relay MACs forwarded packets, and the observed RTT must satisfy (1+δ)·R(s,r) < min_{f∈F}{R(s,f)+R(f,r)}, where the minimum RTT to any point in F is estimated as (3/c)·ShortestDistance(q,F) — fiber-optic links at 2/3 the speed of light. This proof requires only GPS coordinates and local RTT measurements, no BGP modifications or PKI.

2015-levin-alibi §3, §4.4.1 defense

C-Saw's design demonstrates that coupling circumvention capability with censorship measurement creates a self-reinforcing incentive loop: users opt in for improved page load times, their participation grows the vantage-point pool, and richer measurements enable finer-grained technique selection per ISP and URL. The system avoids requiring a pre-populated URL list by building a blocked-URL database dynamically from user-initiated requests.

2015-nisar-case §1 / §3.3 / §4.2 defense

Collaborative spy detection aggregates VPN connection logs (complete, incomplete, and tiny calls) across all volunteer nodes to a central log analyzer, which identifies censor probe IPs by looking for clusters of incomplete or tiny calls from the same /24 block, then distributes a Spy List back to every server so probing packets are silently dropped before the handshake completes. A single server cannot distinguish a spy from a regular client in time; the cross-server aggregate makes pre-response spy identification feasible.

2014-nobori-vpn §4.3 defense

Innocent IP mixing — inserting IP addresses of critical Internet infrastructure (DNS roots, Windows Update servers, popular mail servers) into the relay list distributed to users — forces the censor to manually verify each address before blocking. In March 2013, the GFW blocked every IP VPN Gate mixed in within 30 minutes, demonstrating it was trusting the list without verification; after the technique was noticed (March 20), the GFW switched to verifying IPs first, substantially slowing its blocking cadence.

2014-nobori-vpn §4.2, §6.3 defense

After deploying innocent IP mixing and collaborative spy detection, VPN Gate raised server reachability from China from a low of ~30% to 78.5% by June 19, 2013, sustaining 60–70% reachability through end of August. On August 29, 2013, VPN Gate served 9,000 daily unique IP addresses from China versus Tor's estimated 3,000.

2014-nobori-vpn §6.2, Figure 12 evaluation

An 8-week measurement in June–August 2012 discovered 58,571 unique Freenet installations across 102,376 distinct IP addresses; approximately 25% were in the US and 12.5% in Germany, with Europe and North America collectively representing the vast majority — users from countries typically associated with Internet censorship were a small minority.

2014-roos-measuring §5.1 evaluation

Freenet's deployed Opennet topology uses uniformly random long-range contacts rather than Kleinberg-optimal distance-proportional selection, yielding an average routing length of 37.17 hops in simulation; adopting a 1/d distance distribution (r=1) reduces this to fewer than 13 hops — a 2.9× improvement achievable via a Kademlia-style bucket system.

2014-roos-measuring §4.2 evaluation

Freenet users exhibit a median session length of 95–99 minutes (p=0.975–0.99), substantially longer than all measured P2P file-sharing systems (1–60 minutes for Napster, Gnutella, FastTrack, Overnet, BitTorrent, KAD); ~2% of sessions exceeded 100 hours, and the distribution is best modeled by a lognormal fit (residual error 0.019) rather than Weibull or exponential.

2014-roos-measuring §5.2 evaluation

The paper argues that the advantage in the censor-vs-circumvention arms race lies with the censor due to fundamental asymmetry: a nation state controls centralized communication infrastructure while dissidents depend on it. Standalone anti-censorship tools therefore face a structurally disadvantaged security posture that iterative patching cannot overcome.

2014-tan-censorship §1 defense

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

The paper sketches a decentralized DHT-based communication protocol where all payloads are encrypted in TLS and explicit redirection enables a form of onion routing. Because the censor cannot distinguish censored from non-censored streams, it is forced into a binary choice: block all protocol traffic (overblocking) or allow all of it.

2014-tan-censorship §2.2 defense

If a communication protocol is regularly used for business and commerce, blocking it may be too politically and economically costly for a censor. The paper posits that censorship resistance achieved as a side-effect of widespread general adoption is harder to defeat than a niche protocol designed solely to circumvent censorship.

2014-tan-censorship §1–§2 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

GNS uses a proof-of-work-gated network flood for key revocation, requiring an adversary to block flood traffic on every path between the revocation origin and all peers to suppress it. This is substantially more robust than X.509 certificate revocation lists, which an adversary can render ineffective by simply blocking access to CRL servers — a weakness severe enough that browser vendors must bundle revocation lists inside software updates.

2014-wachs-censorship-resistant §A.5 defense

GNS encrypts all DHT queries and responses using a zone-private-key-derived symmetric key (h = x·l mod n; query = H(hG)) such that a passive DHT observer can only mount a confirmation attack — requiring simultaneous knowledge of both the zone's public key and the specific label. Without both values, an adversary observing DHT traffic cannot determine the label, zone, or record data; even fully participating malicious DHT nodes see only opaque signed blobs unlinkable to their originating query.

2014-wachs-censorship-resistant §4 defense

GNS bounds the trusted computing base (TCB) for any individual name resolution to fewer than approximately 125 entities (constrained by name label length) and makes the full trust chain transparent to the user. By contrast, even simple DNS lookups can silently depend on correct answers from over 100 DNS zones; China's DNS injection caused global collateral damage precisely because out-of-bailiwick NS record chains made the full trust graph invisible to resolvers and users alike.

2014-wachs-censorship-resistant §5 evaluation

Blockchain-based naming systems such as Namecoin are insufficient under a strong adversary model where a nation-state can muster more computational resources than all other participants combined, allowing it to produce alternative valid chain histories. This vulnerability is most acute during system bootstrapping and in censored regions where the user base is small, precisely the conditions under which a censorship-resistant naming layer is most needed.

2014-wachs-censorship-resistant §6 evaluation

TCP-based web traffic performs poorly on mesh networks because each wireless hop halves effective bandwidth (bidirectional ACKs share the same half-duplex channel) and introduces highly variable latency and loss; voice traffic is similarly unsuitable due to jitter. Applications leveraging delay-tolerant networking principles or requiring only very low bandwidth are identified as the category of workloads that can function within mesh constraints.

2013-hasan-building §4.2 evaluation

Purpose-built or uncommon radio hardware provides governments a legal pretext for crackdowns, is subject to import restrictions, and aids identification of dissidents via radio direction-finding equipment. The authors conclude that only ubiquitous, innocuous devices—smartphones and standard indoor WiFi access points—can be used in a dissent network without raising suspicion or endangering users.

2013-hasan-building §5.1 defense

Gupta and Kumar proved that the per-node capacity of a multihop wireless network approaches zero as the number of nodes increases; Li et al. experimentally validated this result for 802.11-based mesh networks. The authors emphasize this is an architectural constraint derived from fundamental radio physics, holding for arbitrary networks regardless of routing protocol.

2013-hasan-building §4.1 evaluation

Mesh networks can reach meaningful scale only by adopting centralized management, planned growth, and a static topology—properties that simultaneously create a single point of failure and make nodes easy targets for government radio direction-finding. Decentralized, organic, mobile mesh retains safety properties but at the cost of near-zero effective capacity as network size grows.

2013-hasan-building §4, §7 defense

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

Because FreeWave is VoIP-provider-agnostic, blocking it requires censors to block all VoIP services simultaneously — a politically and economically costly action given that approximately one-third of U.S. businesses used VoIP by 2011 and penetration was forecast to reach 79% by 2013. The authors argue this collateral-damage cost makes wholesale VoIP blocking infeasible for most censors.

2013-houmansadr-i §V / §II-A defense

MIAB reduces the bootstrap requirement to only the operator's public key — no pre-shared rendezvous point is needed — by using blog pings as a real-time broadcast discovery channel. Since every blog post on the Internet is a potential drop point, the censor cannot enumerate entry points by posing as a legitimate user, unlike Collage (requires an up-to-date task database) or Telex (requires ISP collaboration).

2013-invernizzi-message §3 defense

Key distribution is the primary bootstrapping weakness of steganography-based censorship-resistance systems: a censor can simply block stego-key distribution. Identity-based steganographic tagging (IBST) eliminates this attack surface by requiring only a single master public key, which can be bundled with the client software — no key distribution inside the censored area is necessary.

2013-ruffing-identity-based §1 Introduction / §4 defense

A browser-history survey found that approximately 8% of domain name resolutions involved typing in a genuinely new domain not reachable via an existing link, meaning a SDSI/petname delegation-based name system could serve roughly 92% of real-world Web navigation without requiring any out-of-band key exchange.

2013-wachs-feasibility §4.5 evaluation

In a DHT-based censorship-resistant name system, poisoning attacks (injecting invalid mappings) are neutralized by requiring signature verification on stored values; eclipse attacks (isolating specific mappings from the network) require replication across multiple DHT nodes. Critically, decentralizing lookups from a single ISP resolver to a DHT shifts query visibility from ISPs to arbitrary peers, requiring per-query encryption keyed to secrets known only to the querying client to limit adversaries to confirmation attacks.

2013-wachs-feasibility §4.4 defense

Pseudo-TLDs (e.g., '.key' for cryptographic-identifier namespaces, '.pet' for petname systems) allow multiple censorship-resistant name systems with distinct security trade-offs to coexist transparently alongside DNS via Name Service Switch configuration, with system-specific resolution logic applied per TLD and no application reconfiguration required by users.

2013-wachs-feasibility §4.1 defense

In an adversary model where the censor may hold more computational power than all honest nodes combined, a squatting attack lets the adversary enumerate and pre-register every memorable name, formally proving it is impossible to simultaneously achieve memorable, secure, and global names in a single name system (Zooko's triangle).

2013-wachs-feasibility §3 policy

DEFIANCE's Address-Change Signaling (ACS) requires each client to contact a sequence of IP addresses with precise timing (per-user wait and window parameters) and a one-time passphrase derived from NET provisioning. Connections arriving out of order, outside the timing window, or lacking the correct passphrase receive only innocuous content, so a censor probing a suspected address block finds only normal commodity servers.

2012-lincoln-bootstrapping §4 defense

NET payloads are wrapped in three nested layers — (1) steganographic encoding plus transport encryption with a factory digital signature, (2) proof-of-life (CAPTCHA), and (3) proof-of-work (computational puzzle) — so that even an adversary who harvests many payloads cannot decode them faster than gateway addresses can be rotated. The payload format is explicitly extensible to add harder challenges as adversaries improve.

2012-lincoln-bootstrapping §3.3 defense

The mod_freedom Apache module hooks into the HTTP 404 ErrorDocument handler and steganographically embeds encrypted NET payloads in image responses to valid RP requests, while returning normal content to all other clients. Using Identity-Based Encryption (IBE, Boneh-Franklin) keyed on the server's hostname eliminates any need for out-of-band public-key distribution and allows deployment on thousands of volunteer webservers without mutual trust.

2012-lincoln-bootstrapping §3.1 defense

Encrypted channels expose only two statistical features to an external observer: packet sizes and inter-packet arrival times. Original Traffic Morphing (Wright et al. 2009) shaped only packet-size distributions, leaving inter-packet timing as an unobfuscated fingerprint identical to the source (Tor) distribution. SkypeMorph extends Traffic Morphing to jointly sample from nth-order conditional distributions of both packet sizes and inter-packet delays (tested up to n = 3), closing the timing gap.

2012-moghaddam-skypemorph §4.3, §4.4 defense

The paper explicitly flags that BTP's fixed-size b-byte connection tag creates an active-probing oracle: a censor that sends b−1 bytes and observes no close, then sends one more byte and observes a close, can confirm the endpoint is running BTP. Preventing such active-probing attacks is identified as future work.

2012-rogers-secure §8 detection

BTP's forward secrecy guarantee depends on reliably destroying old keys, but the paper notes that secure deletion from persistent storage—especially solid-state storage—is difficult with current operating systems and hardware. The recommended mitigation is passphrase-derived encryption of stored secrets, though this shifts the problem to passphrase protection.

2012-rogers-secure §6.3 defense

BTP achieves forward secrecy over unidirectional transports—where ephemeral in-band key exchange is impossible—by using a one-way key derivation function (NIST SP 800-108) to produce sequential temporary secrets from an initial shared secret. Once both devices destroy a given temporary secret, no keys derived from it can be reconstructed even if devices are later compromised.

2012-rogers-secure §2, §5.2–5.3 defense

BTP's wire protocol contains no handshakes, timeouts, or plaintext headers. Connections open with a pseudo-random b-byte tag that the recipient can compute in advance from its key state, making BTP frames indistinguishable from random data to a passive observer who does not know the shared secret.

2012-rogers-secure §2, §3.2 defense

BTP's secret retention period for transport t is Rt + 2C + Lt, where Rt is the rotation period, C is the maximum clock-skew tolerance, and Lt is the maximum transport latency. With Rt = 2C + Lt only two temporary secrets need simultaneous storage. Concrete durations: TCP with automatic clocks (C=10s, Lt=60s) requires 2 minutes 40 seconds; TCP with manual clocks (C=1800s) requires 4 hours 2 minutes; mail with manual clocks (Lt=2 weeks) requires 4 weeks 4 hours.

2012-rogers-secure §5.3, §6.2 defense

Twitter's relevance-ranked search returned 53% fewer bot-generated tweets compared to real-time chronological search across 1.1 million queries during the attack; restricting analysis to the top 5 most-recently returned relevance results reduced spam by 64% versus real-time. Relevance ranking incorporates social-graph overlap and content popularity signals to demote mass-produced low-engagement content.

2012-thomas-adapting §5 evaluation

Content-oblivious replication delegates ongoing availability maintenance to 'manifest guarantors' — nodes holding content manifests — who periodically sample chunk replication factors and restore missing replicas without knowing the plaintext they protect, freeing the original publisher from any post-publication obligation. Two honest manifest holders (one content, one key) are sufficient to maintain replication with overwhelming probability even under adversarial conditions and high churn.

2012-vasserman-one-way §3.3 defense

Simulation over erasure code parameters uniformly sampled from m∈[1,5] and n∈[5,500] shows that a 50-of-500 code is the best trade-off between overhead and robustness: it requires nearly 10× storage overhead to support 2^60 variable-size chunks and allows the network to tolerate more than 70% node failure before data is lost. Replication combined with erasure coding yields better durability than either strategy alone.

2012-vasserman-one-way §4, Figure 2 evaluation

A hybrid garbage-collection scheme combining time-based expiry (last-access timestamp cutoff), popularity-based retention, and editor-signed manifest exemptions forces adversaries conducting pollution or exhaustion attacks to continuously re-access or re-upload junk to prevent its deletion. A single honest editor's signature is sufficient to exempt important but infrequently accessed content from deletion indefinitely, while malicious editors cannot explicitly remove content from the system.

2012-vasserman-one-way §3.3 defense

One-way indexing separates a published file into encrypted content blocks (indexed by hash1(block)), a content manifest (indexed by hash2(keyword)), and a key manifest (indexed by hash3(keyword)), so a storer holding all content chunks cannot recover the plaintext or keywords without inverting a cryptographic one-way function. Using distinct hash functions for each manifest type also minimizes the probability that a single node stores both manifests, preventing correlation.

2012-vasserman-one-way §3.1 defense

In a 250-node PlanetLab deployment with 10–15% silent node failures and high churn, the median user retrieved a 20MB file in 65–85 seconds end-to-end (search + manifest download + chunk fetch + reconstruction + decryption). 15.12% of DHT lookups and 11.24% of maintenance operations failed; 20% of nodes accounted for 80% of failures, yet nodes with working connections completed lookups and maintained sufficient guarantors for manifest replication.

2012-vasserman-one-way §4.1, Figure 3 evaluation

Scrambling without secret key management can frustrate DPI-based censors if the de-scrambling function satisfies 'high-inertia' — meaning an adversary computing S⁻¹ on n inputs cannot use less than Θ(n) times the resources of a single commodity-PC user, including electricity, memory, and computation time. This forces bulk censorship to become computationally infeasible without over-censoring all scrambled content.

2011-bonneau-scrambling §1–2 defense

Transmitting the de-scrambling algorithm S⁻¹ as in-page JavaScript alongside AJAX-fetched scrambled content eliminates the need for special client software installation or trusted public-key distribution, removing the primary bootstrapping vulnerability that cryptographic censorship-resistance schemes (including Tor) share — a vulnerability exploited when Iran blocked Tor by filtering its Diffie-Hellman parameter bit sequence.

2011-bonneau-scrambling §1, §3 defense

The proposed multi-stage scrambling composes four orthogonal layers: (a) 128-bit AES with 20 bits stripped, requiring brute-force search; (b) an AES key derived from a CAPTCHA solution; (c) a memory-bound function key; and (d) blocks whose de-scrambling exploits JavaScript floating-point and string-processing quirks. Each layer independently forces a censor to build or emulate a distinct acceleration environment, multiplying total reverse-engineering cost.

2011-bonneau-scrambling §3 defense

Applying a BEAR all-or-nothing package transform (using a zero key) to message blocks forces any censor attempting to scan content to cache all blocks from all active concurrent transfers simultaneously, since no individual block reveals any information about the original message until all blocks are received. Artificially delaying block transmission amplifies censor state requirements proportionally.

2011-bonneau-scrambling §3 defense

By integrating onion routing at the ISP/AS boundary (exactly one onion router per AS hop), the specific relay used is neither specified in the protocol nor addressable by end hosts, eliminating the enumerable relay list that makes overlay Tor blockable. The end host only knows ISP public keys, not individual relay addresses.

2011-liu-tor §5 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

Routing all traffic through rendezvous mailboxes with a put/get interface prevents direct DoS against end hosts and hides service host addresses; each service deploys many mailboxes with a randomly chosen sequence per connection, so an attacker can disrupt at most a fraction of any given flow even with substantial resources.

2011-liu-tor §5, §7.1 defense

The design guarantees that as long as an end host can reach any non-censoring ISP, it can trampoline to any service; the anonymity properties make it difficult for ISPs to selectively block flows without cutting off the end host from the outside world entirely. Wikileaks-like services require only one willing authority for name resolution, not universal cooperation.

2011-liu-tor §7.1 defense

Channel blocking risk in Proximax is modeled as an independent Poisson process with rate λj; when a proxy is advertised on multiple channels simultaneously the risk parameters add (Λi = γ + Σλj), so each additional dissemination channel shortens expected proxy lifetime 1/Λi. The analytic result is that redundant multi-channel broadcasting is strictly suboptimal once cumulative risk exceeds the marginal usage gain.

2011-mccoy-proximax §3.1 defense

A sophisticated censor can infiltrate a proxy distribution system, accumulate large numbers of proxy addresses and channel identities, and delay mass-blocking for weeks or months to maximize information before acting. The paper argues this is self-limiting: delayed blocking extends proxy lifetimes (benefiting system yield), and the infiltrating account's subtree reputation score degrades sharply the moment it begins blocking proxies, triggering exclusion from future proxy assignments.

2011-mccoy-proximax §4 detection

Open proxy distribution registrations are vulnerable to adversary flooding with fictitious accounts that inflate yield scores via dummy connections. Proximax uses invitation-only registration with RICO-style subtree reputation scoring — a compromised sub-node taints the entire inviting user's subtree — and sub-linearly credits usage from closely clustered source IP prefixes to limit bot-driven inflation.

2011-mccoy-proximax §2.2, §4 defense

Proximax frames proxy distribution as a yield-maximization problem: the expected yield of a proxy is its attracted usage Ui divided by its total blocking risk Λi. A dissemination channel should only be assigned a proxy if the channel's own yield ratio u/λ exceeds the proxy's current yield ratio; otherwise the added risk outweighs the additional traffic and the channel must not be used at all.

2011-mccoy-proximax §3.3 defense

Each round of copyright enforcement drove deeper architectural decentralization: centralized servers (BBSs/FTP) → central directory (Napster) → supernodes (KaZaA/Grokster) → pure protocol (BitTorrent). Even after Grokster was shut down its software continued to work, because no fixed corporate entity remained as the control point.

2011-seltzer-infrastructures §1.1 defense

Censorship operating at the infrastructure layer (hosting, DNS, ISPs) rather than the content layer produces opacity: blocklists must be kept secret lest they become menus of blocked content, accuracy cannot be examined, and harms are divided from those with incentive or expertise to oppose them. The consistent pattern in anti-censorship responses is to distribute, decentralize, encrypt, and obfuscate — making circumvention traffic indistinguishable from permitted use.

2011-seltzer-infrastructures §2.1 policy

Rateless erasure coding with ε=0.01 adds only a 0.5% storage and traffic overhead. Consistent hashing of message identifiers to task-database entries ensures that when 50% of tasks are replaced, sender and receiver still share at least one task if three or more tasks are mapped per identifier. At a 10× send rate, message recovery succeeds even if 90% of published vectors are blocked.

2010-burnett-chipping §4.2, §4.3, Figure 4, Figure 5a defense

The paper demonstrates that no single steganographic algorithm can provide both availability and deniability, since almost all production algorithms have been broken and steganography alone does not hide the identities of communicating parties. Collage addresses this by treating the embedding algorithm as a swappable component in a layered architecture—vector layer, message layer, application layer—so that compromise of the embedding scheme does not compromise the system, and stronger algorithms (e.g., digital watermarking) can be substituted as they mature.

2010-burnett-chipping §4.1, §2 defense

A dynamic binary-tree partitioning algorithm solves the proxy distribution problem with at most k(1 + ⌈log₂(n/k)⌉) total proxy keys: partition n users into k groups in round 1, then halve each compromised group on each compromise event. Each of k adversaries can trigger at most ⌈log₂(n/k)⌉ compromises, bounding total proxy expenditure tightly.

2010-mahdian-fighting §4 Theorem 3 defense

A simple entropy argument proves the dynamic key distribution problem requires at least Ω(k log(n/k) / log(k + log n)) keys: the algorithm must identify which k of n users are adversaries from at most ℓ log ℓ bits of feedback (ℓ round outcomes each indexing one of ℓ keys), and distinguishing among C(n,k) adversary sets requires log C(n,k) = Ω(k log(n/k)) bits.

2010-mahdian-fighting §4 Theorem 6 evaluation

The static proxy distribution problem — giving k²-adversarial users keys from m proxies so that all n−k legitimate users retain at least one uncompromised proxy — requires at most O(k² log n) keys and cannot be solved with fewer than Ω(k log(n/k)) keys. This establishes the information-theoretic cost of one-shot proxy distribution against k colluding informants among n users.

2010-mahdian-fighting §3 Theorem 2 defense

By reusing keys already held by trusted (non-suspicious) users for ℓ−1 of ℓ subgroups when bisecting the suspicious cohort — issuing only one fresh key per round — the total proxy count drops from O(k log n) to O(k² log n / log log n) in expectation. The information-theoretic lower bound is Ω(k log(n/k) / log(k + log n)), so this bound is tight in n up to a factor of k.

2010-mahdian-fighting §4 Theorems 4–6 defense

In invitation-based proxy networks (modeled on Psiphon's trust-tree), a single adversary can invite fake accounts as children in the trust tree, multiplying the effective adversary count k and invalidating sublogarithmic key budgets. For k=1 adversary on a trust tree of depth O(log n), an O(log n)-key algorithm exists by keeping the 'suspicious group' always rooted at a subtree boundary; for k>1 this remains an open problem.

2010-mahdian-fighting §5 Theorem 7 defense

In the Clouds P2P protocol, a blocking attack against a specific topic requires adversaries to occupy at least 50% of the 200-peer region closest to the resource provider to be effective; below that threshold, query messages routed through multiple paths bypass the censorship. This 50% threshold holds regardless of the number of clouds κ created per peer.

2009-backes-anonymity §5.4 evaluation

The number of clouds per peer κ has no measurable effect on censorship resistance (Figure 5 curves are identical across κ = 1–4), while cloud size is the dominant driver of message overhead. This decoupling means designers can increase κ to improve anonymity without degrading censorship resistance or incurring bandwidth cost.

2009-backes-anonymity §5.4–§5.5 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

The Clouds protocol retrieves approximately 70% of available answers even in the absence of attackers, representing a ~30% retrieval performance decrease relative to an insecure SON. This baseline loss stems from the cloud-based routing mechanism's probabilistic message delivery, not from adversarial interference.

2009-backes-anonymity §5.6 evaluation

Kaleidoscope uses at most one intermediate relay hop so proxies can serve users beyond their immediate trust neighborhood without directly learning user addresses. If a system allowed each proxy to directly advertise to N users, a censor posing as a proxy would learn N user identities; the one-hop relay design caps per-proxy exposure to r=5 relay addresses and keeps end-user identities hidden from proxies.

2008-sovran-pass §3 defense

On a crawled Orkut subgraph of 42,474 users (≈90% Brazilian nodes treated as the censored domain, 15% of external nodes as proxies = 1.5% overall), the median node reaches 7 proxies — higher than the synthetic graph due to greater average degree (5.59 vs. 4.65) and lower clustering. Even when subverted trust links reach half the total proxy count, more than 94% of users can still access at least one proxy unknown to the censor.

2008-sovran-pass §3.2 evaluation

Kaleidoscope bounds censor knowledge by routing proxy advertisements over symmetric random routes of length r=5 on a social trust graph: if the censor controls f subverted trust links, they can learn of at most r×f = 5f users or proxies regardless of how many Sybil identities they generate. Symmetric routing ensures the set a node learns of and the set that learns of a node are identical, closing the asymmetric information-leakage channel.

2008-sovran-pass §3, §3.2 defense

Simulation on a synthetic social graph of one million nodes (average degree 4.65, maximum 13) shows that when 1.5% of nodes act as proxies and random routes of length r=5 are used, the median node can reach 3 proxies and more than 90% of nodes can access at least one proxy.

2008-sovran-pass §3.2 evaluation

When the GFC keyword blacklist is known, multiple server-side-only evasion techniques become viable requiring no client modification: IP packet fragmentation to split keywords across MTU boundaries, HTML comment injection mid-keyword (e.g., 'Fa<!- Comment ->lun Gong'), alternative URL percent-encodings (e.g., 'F%61lun Gong'), and spam-style character substitution ('F@1un G0-ng'); the GFC implementation was observed not to check control characters in URL requests.

2007-crandall-conceptdoppler §5 defense

Theorem 1 proves that censorship resistance (CR) implies Private Information Retrieval (PIR): any system achieving low censorship susceptibility must implement PIR as an underlying primitive. CR systems cannot be built with cryptographic primitives weaker than PIR.

2005-perng-censorship §5 defense

Server-deniability schemes (Publius) and data-entanglement schemes (Tangler, Dagster) both achieve censorship susceptibility of 1 under the cooperative-server model. Publius fails because the Publius URL encodes the hosting servers and document identity in public, enabling direct query filtering. Tangler and Dagster fail because their limited-width entanglement graphs allow a censor to remove a document with collateral damage too small to prevent selective censorship — only a small number of blocks per document are entangled.

2005-perng-censorship §4.3–4.4 evaluation

PIR alone does not achieve censorship resistance. Using the QRA (Quadratic Residuosity Assumption) PIR scheme as a direct CR implementation, a filter can replace a query component — substituting a quadratic residue for a non-residue at the target column index — forcing the server to return an incorrect result for the targeted document while leaving all other documents unaffected, yielding censorship susceptibility of 1.

2005-perng-censorship §6 detection

Theorem 3 demonstrates that having the server digitally sign its response together with the verbatim client query is sufficient to achieve CR when built atop any secure PIR protocol. This construction (sys+S) eliminates query modification as an attack vector, reducing the censor's viable strategies to query-dropping only — an advantage bounded above by the underlying PIR adversary's advantage, proving that the censor must shut down the entire service to achieve selective filtering.

2005-perng-censorship §7 defense

Under a threat model granting the censor universal inspection of server communications and processing logs — with only the server's signing key withheld — data-replication systems (Freenet, Gnutella, Eternity Service) and anonymous-communication systems (Free Haven, Serjantov's scheme) all achieve censorship susceptibility of 1. Because document names are publicly known, a censor with full server visibility can selectively drop any targeted query without disrupting access to other documents.

2005-perng-censorship §4 detection

The paper argues that censorship is an economic activity in which both censor and target incur costs, and that binary 'blocked/unblocked' models are as unrealistic as an omnipotent global adversary. Technology changes (e.g., moveable type, online publishing, trusted computing) can shift the cost parameters dramatically, making quantitative cost modeling — rather than binary vulnerability analysis — the correct framing for censorship-resistance evaluation.

2004-danezis-economics §6 policy

Discretionary P2P networks avoid the social-choice and incentive-manipulation problems inherent in random distribution, which requires collective agreement on a system-wide resource ratio (rs, bs) and thus creates incentives to subvert voting or reputation mechanisms. By allowing nodes to self-select content, discretionary systems need no election schemes, reputation systems, or electronic cash, enabling simpler and more stable designs.

2004-danezis-economics §7 defense

Under the paper's economic model, the aggregate censorship-resistance defense budget is always at least as large in a discretionary P2P network (nodes serve content they choose) as in a random-distribution network: for every node i, td ≥ ts, so the total cost imposed on the censor satisfies Σtd ≥ Σts. Equality holds only when all nodes share identical preferences (ri = rs); in all other cases discretionary distribution is strictly harder to censor.

2004-danezis-economics §5 defense

In a random-distribution network, nodes whose utility is non-decreasing under censorship will set their defense budget to zero. For example, in a network with rs = 0.5 (equal red/blue), a censor shifting the distribution to rc = 0 (all blue) increases the utility of strongly blue-preferring nodes; they then invest nothing in resistance, reducing aggregate network defense.

2004-danezis-economics §4–§5 defense

Under the paper's quadratic utility function and linear defense probability P(t) = t/T, a node will invest zero resources fighting censorship when the censor's imposed distribution reduces its utility by less than half (i.e., when Ui(rc,bc) ≥ Ui(ri,bi)/2). Nodes whose preferences most diverge from the censor's are the first to resist; mild censorship therefore attracts little aggregate resistance.

2004-danezis-economics §6.2 evaluation

The paper presents a systematic taxonomy of blocking criteria across ISO/OSI layers: circumstance-based (addresses including sender/receiver/kind/physical location; timing including send time, receive time, duration, frequency; data-transfer properties; services including protocols, names, addresses) and content-based (file type/MIME, statistical detection of encrypted or compressed data, pattern matching for keywords or phrases, and website fingerprinting via request-count/byte-volume signatures).

2004-k-psell-achieve §4, Appendix A defense

The paper proposes using CAPTCHAs (hard AI problems) to gate forwarder-list access, forcing the blocker to expend human resources solving every puzzle while each blockee solves only one. However, a 'stealing cycles from humans' attack allows a censor to relay CAPTCHAs to unwitting third parties (e.g., visitors to an attacker-operated website) who solve them on the censor's behalf.

2004-k-psell-achieve §5.3 defense

The protocol between blockee and volunteer forwarder is designed to be transport-layer independent from the outset, allowing substitution of plain TCP with SSL tunnels, SMTP, or steganographic channels as the censor escalates detection. The system is intentionally deployed in a weak initial form to observe how quickly and in what manner the censor adapts, then hardened iteratively based on measured censor behavior.

2004-k-psell-achieve §6.2 defense

Active-server document anonymity is achieved by routing decryption through a randomly chosen ephemeral 'decrypter' node: the storer holds only ciphertext {h}k while key k is delivered separately to the decrypter via onion routing. Neither the storer nor any other single node can reconstruct the plaintext share, so a storer cannot identify the document it is hosting even by attempting to retrieve it.

2002-serjantov-anonymizing §2–§3 defense

An adversary who wishes to expose storers by having forwarders log storer identities must compromise all n−k+1 chosen forwarders before or during the publication event; forwarders that legitimately delete the storer mapping immediately after acknowledging publication render this attack ineffective unless the adversary pre-positions malicious nodes at sufficient density. The paper notes that with a reasonably large forwarder population the probability of the required simultaneous compromise is small.

2002-serjantov-anonymizing §3 defense

The paper proposes a forwarder/storer role split in which forwarders hold only an anonymous return-address pointer to the storer, and deliberately forget the storer's identity upon receiving a publication acknowledgment. Because forwarders neither hold content nor retain storer addresses post-publication, coercing a forwarder after publication yields no actionable information about where shares are held.

2002-serjantov-anonymizing §2–§3 defense

Publius splits document keys into n shares where any k reconstruct the document, requiring a censor to coerce only n−k+1 servers to suppress it. Because all Publius server locations are discoverable by any reader, the paper argues this threshold is easily achievable, making location-secrecy of storers a necessary — not optional — property for censorship-resistant storage systems.

2002-serjantov-anonymizing §1 defense

TCP RSTs are delivered unreliably and different OS stacks apply different validity rules, so a NIDS cannot safely tear down connection state on RST alone; a 'reliable RST' scheme — sending a keep-alive ACK behind every forwarded RST and tearing down state only upon observing a confirming RST from the trusted side — resolves this without violating end-to-end semantics. The cold-start problem (state loss on restart) can be addressed statelessly by stripping payload from unknown-connection packets from untrusted hosts and probing the trusted endpoint with a keep-alive before instantiating state.

2001-handley-network §6.1–6.2 defense

A traffic normalizer placed inline ('bump in the wire') can eliminate over 70 IP/TCP packet-level ambiguities before a NIDS inspects traffic — including fragment reassembly, TTL restoration, DF flag clearing, IP option removal, and cryptographic IP ID scrambling — leaving the classifier with an unambiguous byte stream and removing the degrees of freedom an attacker needs to evade detection.

2001-handley-network §3, Appendix A defense

The paper derives a closed-form expression for the expected number of later blocks that link to the n-th block: with c=10 cross-links per block, there is a 55% probability that the 10^7th block in the system will have been linked by at least one subsequent legitimate block after 10^5 additional blocks are added. This quantifies the minimum corpus activity required before a publisher can safely announce a document and have plausible censor-resistance.

2001-stubblefield-dagster §5.5 evaluation

Dagster identifies every block by the cryptographic hash of its contents (block ID), making it infeasible for an adversary to pre-empt a name with bogus data — an attack that directly affects Publius, where an attacker who possesses a target document can insert garbage under the same name that the legitimate document would have occupied. Content-addressing also makes the system robust to the naming ambiguity observed in Freenet (where a single document was posted under three distinct capitalizations).

2001-stubblefield-dagster §4.3, §5.2 defense

Dagster achieves censorship resistance on a single server — without geographic replication — by cryptographically intertwining legitimate and illegitimate data into a directed acyclic graph. Each new block XORs the publisher's content with c pre-existing blocks before encrypting with a fresh key, so removing any one block destroys the decodability of every block that later links to it. This creates a legal constraint: a censor cannot excise a censorable block without simultaneously destroying an unknown number of legally protected blocks that depend on it.

2001-stubblefield-dagster §1, §3.1, §5.5 defense

Dagster's randomness predicate cannot distinguish legitimate random-looking blocks from adversarially generated filler, leaving the system vulnerable to storage-exhaustion denial-of-service: an attacker can submit arbitrarily many random blocks that pass the predicate, consuming server disk until legitimate publications are refused. The paper identifies anonymous digital cash (as proposed in the Eternity Service) or hash-cash proof-of-work as candidate mitigations but does not implement either.

2001-stubblefield-dagster §7 defense

Publius cryptographically binds the URL to both the document content and the key shares via name_i = wrap(H(M · share_i)). Any unauthorized modification to the stored encrypted file, a share, or the URL itself causes the tamper check to fail, preventing silent content substitution by a malicious server.

2000-waldman-publius §3.2, §3.3 defense

A malicious server operator with write access can mount a redirection attack by inserting a fake update file pointing to adversary-controlled content. If the client retrieves only k shares and Mallory controls k collaborating servers, all k update URLs match and the client proxy follows the redirect. A 1-bit non-updatable flag in the Publius URL blocks this vector by instructing clients to ignore all update files.

2000-waldman-publius §5.2 evaluation

Publius's delete mechanism requires the publisher to supply H(server_domain · PW) per server rather than a bare password, preventing any single malicious server from learning the global password and deleting the document from all hosting servers. However, the paper acknowledges that an adversary who identifies the publisher can apply coercive ('rubber-hose') methods to obtain the URL and password directly from the author, bypassing all cryptographic protections.

2000-waldman-publius §3.4, §5.5 defense

Publius provides source anonymity once content is published but offers no connection-based anonymity at upload time. A network-layer eavesdropper between the publisher and the servers, or a server's connection log, can reveal the publisher's IP address. The paper explicitly states that Publius must be combined with a mix-network or crowd-anonymity tool (e.g., Crowds, Onion Routing) to protect publisher identity during the upload phase.

2000-waldman-publius §5.4 evaluation

Publius encrypts content under a symmetric key K, then splits K into n shares using Shamir secret sharing such that any k shares reconstruct K. Each server stores the encrypted document plus one share, so an adversary must corrupt or destroy n−k+1 servers to censor the document, and increasing n or decreasing k raises the bar proportionally.

2000-waldman-publius §3.2, §5.1 defense

The paper proves that any network IDS operating without maintaining complete, OS-specific per-connection state cannot reliably reconstruct the byte stream seen by the end-system. TCP and IP reassembly ambiguities guarantee unavoidable blind spots unless the IDS performs full per-target OS emulation—a fundamental architectural limitation, not an implementation bug, that applies equally to any DPI-based censor.

1998-ptacek-insertion §5 defense

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

Using Byzantine-fault-tolerant protocols (specifically Rampart), seven replicas suffice to resist a conspiracy of any two malicious administrators or the accidental destruction of four systems with guaranteed complete recovery. Signing all files with a system key further ensures that a full recovery is possible as long as a single valid copy and an uncompromised public key survive.

1996-anderson-eternity §4.6 defense

Effective censorship of a distributed service requires simultaneous enforcement across every jurisdiction hosting nodes. With no head office to coerce, a legal attack requires coordination across multiple independent legal systems — making successful suppression 'very expensive indeed — hopefully beyond even the resources of governments.' Local bans (e.g., country-level) do not affect nodes in other jurisdictions.

1996-anderson-eternity §4.1–§4.2 policy

The Eternity Service's core design stores a file on 100 servers worldwide but retains records of only 10 for auditing, destroying the remaining 90 records. Even if a user is legally compelled to disclose all 10 known server locations and those copies are seized, 90 copies survive at unknown locations and can be retrieved via anonymous broadcast once the user leaves the jurisdiction.

1996-anderson-eternity §4.3 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