wifi-densepose/examples/research-sota
rUv db64b4c671
research(R3): cross-room re-ID — MERIDIAN closes the env-shift gap + 4 privacy constraints (#715)
Synthesis of AETHER (ADR-024) + MERIDIAN (ADR-027) + privacy framing
+ identified next research lever (physics-informed env prediction).

Simulation results (10 subjects, 3 rooms, 128-dim embeddings, env/person
scale ratio 4.7x):

| Configuration                            | 1-shot acc |
|------------------------------------------|-----------:|
| Within-room (matches AETHER ~95% target) |      100%  |
| Cross-room, raw cosine K-NN              |       70%  |
| Cross-room, MERIDIAN 100% env removal    |      100%  |
| Cross-room, MERIDIAN 70% env removal     |      100%  |
| Chance                                   |       10%  |

The 30 pp gap from within-room to raw cross-room is the angular
contribution of env-shift that cosine similarity can't normalise away.
MERIDIAN per-room centroid subtraction recovers it -- robust even at
70% effectiveness (realistic for limited labelled examples).

Privacy framing: R14 baseline + 4 new constraints specific to
biometric-class re-ID data:
1. No cross-installation linkage
2. Embedding storage requires explicit opt-in (biometric consent class)
3. Cryptographically verifiable forgetting
4. No re-ID across legal entities

These rule out cross-building tracking, mass surveillance, long-term
unlabelled storage, third-party sharing. They allow per-installation
personalisation, household anomaly detection, multi-person pose
association in the same room.

R3 closes the loop on R14's empathic-appliance vision: re-ID is THE
primitive that makes per-occupant features possible. Without R3,
R14's verticals can't ship.

Identifies next research lever: physics-informed env_sig prediction
from R6's forward operator + room map = zero-shot cross-room transfer
without labelled examples in the new room.

Composes:
- R5/R6: person+env decomposition in embedding space
- R7: mincut = defence against re-ID spoofing
- R9: RSSI K-NN showed env-locality dominance for the K-NN primitive
- R14: 4 new constraints extend R14's framework to biometric class

Honest scope: additive decomposition is first-order; real CSI env
effects are multiplicative in subcarrier domain. Adversarial scenarios
not simulated.

Coordination: ticks/tick-12.md, no PROGRESS.md edit.
2026-05-22 02:13:10 -04:00
..
r1_toa_crlb.py research(R1): ToA CRLB — precision floor for WiFi multistatic localisation (#711) 2026-05-22 01:38:35 -04:00
r1_toa_crlb_results.json research(R1): ToA CRLB — precision floor for WiFi multistatic localisation (#711) 2026-05-22 01:38:35 -04:00
r3_crossroom_reid.py research(R3): cross-room re-ID — MERIDIAN closes the env-shift gap + 4 privacy constraints (#715) 2026-05-22 02:13:10 -04:00
r3_reid_results.json research(R3): cross-room re-ID — MERIDIAN closes the env-shift gap + 4 privacy constraints (#715) 2026-05-22 02:13:10 -04:00
r5_subcarrier_saliency.py research(sota): kick off SOTA research loop + first R5 saliency measurement (#702) 2026-05-21 23:05:55 -04:00
r6_fresnel_results.json research(R6): Fresnel-zone forward model — bedrock physics for CSI sensitivity (#710) 2026-05-22 01:31:09 -04:00
r6_fresnel_zone.py research(R6): Fresnel-zone forward model — bedrock physics for CSI sensitivity (#710) 2026-05-22 01:31:09 -04:00
r7_multilink_consistency.py research(R7): Stoer-Wagner mincut detects adversarial CSI nodes 3/3 in synthetic (#704) 2026-05-21 23:28:46 -04:00
r7_multilink_consistency_results.json research(R7): Stoer-Wagner mincut detects adversarial CSI nodes 3/3 in synthetic (#704) 2026-05-21 23:28:46 -04:00
r8_rssi_only_count.py research(R8): RSSI-only person count retains 95% of full-CSI accuracy (#703) 2026-05-21 23:18:09 -04:00
r8_rssi_only_results.json research(R8): RSSI-only person count retains 95% of full-CSI accuracy (#703) 2026-05-21 23:18:09 -04:00
r9_rssi_fingerprint_knn.py feat(tools/ruview-mcp): M2 — wire real inference via cog health (#706) 2026-05-21 23:43:32 -04:00
r9_rssi_fingerprint_results.json feat(tools/ruview-mcp): M2 — wire real inference via cog health (#706) 2026-05-21 23:43:32 -04:00
r10_foliage_attenuation.py research(R10): through-foliage wildlife sensing — physics feasibility + per-species gait taxonomy 2026-05-22 00:59:11 -04:00
r10_foliage_results.json research(R10): through-foliage wildlife sensing — physics feasibility + per-species gait taxonomy 2026-05-22 00:59:11 -04:00
r11_maritime_propagation.py research(R11): maritime sensing — through-bulkhead impossible, through-seam works (#712) 2026-05-22 01:53:51 -04:00
r11_maritime_results.json research(R11): maritime sensing — through-bulkhead impossible, through-seam works (#712) 2026-05-22 01:53:51 -04:00
r12_rf_weather_eigenshift.py research(R12): RF weather mapping eigenshift — negative-ish, with clearly-actionable revision path (#707) 2026-05-21 23:52:49 -04:00
r12_rf_weather_results.json research(R12): RF weather mapping eigenshift — negative-ish, with clearly-actionable revision path (#707) 2026-05-21 23:52:49 -04:00
r13_bp_physics_floor.py research(R13): NEGATIVE — contactless BP from CSI is physically inferior to a cuff (#713) 2026-05-22 02:00:35 -04:00
r13_bp_results.json research(R13): NEGATIVE — contactless BP from CSI is physically inferior to a cuff (#713) 2026-05-22 02:00:35 -04:00