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).

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.

§4.4, Table 5, Table 6 evaluation

State-of-the-art ML classifiers (Deep Fingerprinting, Decision Tree, Random Forest, nPrintML) trained on known UPGen protocols and benign traffic always incur high out-of-distribution false-positive rates when attempting to block unknown UPGen protocols — in the vast majority of experiments the OOD FPR is 100%. The one exception (SSH OOD, Deep Fingerprinting) achieved a UPGen TPR of only 20%. By contrast, identical classifiers successfully generalize to block unknown Obfs4 flows with near-zero collateral damage in 3 of 4 cases.

§4.3.3, Table 3, Table 4 evaluation

In laboratory benchmarks, the best UPGen-generated protocol achieves 252 ms TTFB latency (vs 212 ms Obfs4, 313 ms TLS) and 4.25 Gbit/s throughput per core (vs 4.65 Gbit/s Obfs4, 9.42 Gbit/s TLS). The worst-case UPGen protocol (4.5 RTT handshake) reaches 677 ms TTFB but 3.70 Gbit/s throughput. In large-scale distributed Tor simulations, the choice of UPGen protocol had no statistically significant effect on end-to-end Tor flow performance.

§5, Table 7 evaluation

Stage 1 of the detection pipeline uses a lightweight heuristic: restrict analysis to IP addresses in "VPS-dense ASNs," which censors already target for resource-intensive inspection of fully-encrypted traffic. This pre-filter dramatically reduces the search space before applying the more expensive dual-role behavioral analysis. The evaluation was conducted without Stages 1 and 3 due to dataset limitations, meaning the reported 23% recall and 0.18% FPR are conservative lower bounds on the full pipeline's performance.

2026-almutairi-server §2.1, §2.2, §3.4 fully-encrypted-detectdpiml-classifier

Adversarial pre-padding — prepending stochastic byte noise to packets — degrades ET-BERT encrypted traffic classification accuracy from >99% to 25.68%, exposing a structural vulnerability in all payload-byte-dependent detection systems. White-box adversarial attacks (Ayaka AH-MSI) additionally achieve evasion rates exceeding 99.5% against standard continuous-time sequence models via Manifold Shattering, where adversaries align malicious temporal distributions with benign baselines.

2026-ferrel-aegis-adversarial-entropy-guided §II, §VI-A dpiml-classifier

Explicitly disentangling packet headers (structured, low-entropy) from encrypted payloads (high-entropy, stochastic) into separate MoE branches yields consistent gains across six datasets: 86.85% F1 on 120-class TLS 1.3 traffic (CSTNET-TLS), 97.88% F1 on USTC-TFC2016 malware/benign flows, and 92.65% F1 on imbalanced IoT traffic (CIC-IoT2022), demonstrating that headers and payloads carry fundamentally different and independently exploitable discriminative signals.

2026-he-trafficmoe-heterogeneity-aware-mixture §III-A, §IV-B, Tables II–III ml-classifierdpifully-encrypted-detect

Striding with factor 4 (early downsampling) produces the largest single-factor degradation in the ablation study: average macro-F1 drops from 0.9909 to 0.9772 and cross-dataset variance increases from 4.77×10⁻⁵ to 4.51×10⁻⁴, with worst-case dataset performance falling to MIN 0.9524. Fine-grained byte order and short-range structure — protocol headers, payload signatures, repeated byte motifs — carry essential discriminative signal that stride-based aggregation destroys.

2026-kulatilleke-mambanetburst-direct-byte-level §V-B, Table IV ml-classifierdpi

A burst of just 5 packets truncated to 320 bytes each (1600 bytes total) suffices for macro-F1 ≥0.9824 across all six benchmarks; the classification token reads from the final recurrent state after a 4-layer Mamba-2 stack processing this fixed-length prefix, with no additional flow-level or session-level context required.

2026-kulatilleke-mambanetburst-direct-byte-level §III-A, §V-A ml-classifierdpi

Classification from the first 5 packets × 320 bytes (1600-byte burst) achieves near-perfect accuracy across Tor (F1=0.9990), VPN (F1=0.9871), malware (F1=0.9954), and IoT attack traffic (F1=0.9966), with IP addresses masked and only header and initial payload retained. The earliest portion of each packet provides sufficient discriminative information for a classification decision made within the first kilobyte of a flow.

2026-kulatilleke-mambanetburst-direct-byte-level §III-A, §V-A, Table II–III ml-classifierdpi