 Claude
|
192ed2a236
|
feat(privshield): implement SOTA-driven attackers and compliant controls (ADR-288 §sota)
From the verified 2025-2026 deep-research findings, all four approved code items,
each opt-in so the reference witness stays byte-identical (0x350d…f448):
- attacker: BFI->CSI Reconstruction adversary (BFIAttack) — recovers the
direction of the CSI consistent with the *captured* report; a secret
orthogonal rotation leaves it at chance (no key to invert). AdaptivePooling
adversary (PrivISAC) — pools + whitens per identity; still collapses.
`AttackerKind` selects the shape.
- protector: `ObfMode::PerPacketUnitary` — fresh per-packet unitary, AP-side and
client-transparent (LeakyBeam family). `dp_epsilon` — ε-DP angular dither,
renormalized to preserve emission energy (still not jamming).
- throughput: `dp_residual` makes ε a real privacy<->throughput knob (smaller ε
costs more gain).
- experiment: `attacker_kind` + mode-aware keying dispatch.
Tests (43 pass, +5): reconstruction & adaptive-pooling collapse (and win
unprotected); per-packet mode collapses + compliant; ε-DP still collapses +
compliant; DP throughput frontier monotonic. clippy -D warnings clean, fmt
clean, wasm --lib builds. All new numbers remain SYNTHETIC/L0.
Docs: 09-sota-update backlog items 1-4 marked implemented; crate README module
table refreshed.
Co-Authored-By: claude-flow <ruv@ruv.net>
Claude-Session: https://claude.ai/code/session_01WEXNqzs7UsfNFBcP5yW21p
|
2026-08-09 16:20:18 +00:00 |
|
|
Claude
|
c63b26034b
|
docs(privshield): fold verified 2025-2026 SOTA into threat model, compliance, and roadmap
From a fan-out deep-research run (20 primary sources, 25 claims 3-vote verified,
24 confirmed / 1 refuted):
- New docs/research/privacy-shield/09-sota-update-2026.md: cited, evidence-classed
SOTA update + prioritized VEIL improvement backlog.
- ADR-288 gains a "2025-2026 evidence update" section: broader threat (BFId
99.5%/N=197; LeakyBeam through-wall vitals @20m; WiKI-Eve/SThief keystrokes;
BFIAttack BFI->CSI reconstruction), VEIL's family independently validated
(LeakyBeam per-packet unitary 89.7->51%; PrivISAC RIS 93->30%), BeamDancer
(IEEE TWC 2024) as compliance precedent, shield-security-is-CLAIMED honesty,
and the unfilled governance gap. Do NOT cite BeamDancer's refuted >96% PDR.
- Roadmap §3.1: answers "does this need custom WiFi firmware?" — yes; ESP32 is an
attacker/sensor node only (closed blob, CSI read only), the protector needs
openwifi / Nexmon / vendor firmware; keyed-reversible needs both ends + key.
Docs only. All VEIL numbers remain SYNTHETIC/L0; no code or claims upgraded.
Co-Authored-By: claude-flow <ruv@ruv.net>
Claude-Session: https://claude.ai/code/session_01WEXNqzs7UsfNFBcP5yW21p
|
2026-08-09 16:09:33 +00:00 |
|