Asking the fleet what it is doing…
monad-knowledge Wi-Fi sensing lab · FIIT STU
Campaign session

01KY0RR3SS30H1P6NP9QXM434Y

finished 2026-07-20 22:01:07.641670+00:00 → 2026-07-20 22:02:17.747788+00:00 · 6 runs · supervisor: react-agent

“Real S3DIS geometry made crowd-walkable (osm_clean tooling); 3 furnished floors, coupled crowd+CSI+device-traffic on Metal. Goal 2 (flow&dwell atlas) delivered: 244 agents, 22,482 WiFi emissions, published Artifact. Goal 1 (within-space device-free counting R²): area1 0.64, area5 0.05, area6 0.01 — weak, floor-dependent. Goal 3 (cross-space transfer R²): all negative (−3 to −67) — no transfer. HEADLINE: device-based CSI is occlusion-killed in walled buildings (3/7128 links reach APs); device-free counting works only where APs overlap occupied rooms. HONEST SELF-CRITIQUE: the building-wide counting framing was naive (1 central Tx → scattered APs across walls is a strawman); the real result is that CSI counting is inherently PER-ROOM and occlusion is the defining feature, not a bug. The valuable reframe (per-room counting + sensor-placement coverage + BLE room-assignment fusion) is the recommended next campaign. Sim-only; real AX210/Pi5 walled floor (IP-106/IP-112) pending.”

Archive snapshot, as of 1 day ago — the run corpus is rebuilt once a day, so this page is not a live reading. The fleet panel is the live one; it refreshes every 30 s.

Success criteria

CriterionResolved
Goal 2: flow & dwell showcase (heatmaps/trajectories/dwell + device traffic) on real geometry yes
Goal 1: per-space CSI counting fidelity yes
Goal 3: cross-space counting transfer yes
Real geometry (no synthetic), furnished, crowd-walkable yes
Framing: sim-only hypothesis generator, honest about limitations yes

Synthesis

Real crowds in real buildings — what worked, what didn't, and the reframe

Campaign: c-public-spaces · Session: 01KY0RR3SS30H1P6NP9QXM434Y · Substrate: S3DIS (real scanned Stanford building, areas 1/5/6) · 6 coupled runs on Metal

Executive summary

We took a real scanned building — real rooms, real furniture, no invented geometry — made it crowd-walkable, and ran real crowds whose phones emit WiFi traffic into a ray-traced channel, across three floors. The flow & dwell showcase works (an interactive atlas of 244 agents and 22,482 device emissions). The crowd-counting results are honestly weak: within a floor, WiFi counts the crowd well on one floor (R²=0.64) and barely on the others (0.05, 0.01), and a counter trained on one floor does not transfer to another. The reason is physical and, in hindsight, the whole point: walls block the radio, so a ceiling access point mostly senses its own room. Counting a walled building from a few scattered sensors is the wrong question — and that realisation is the campaign's real output.

What we built (the durable value)

The hard part was infrastructure: real dataset floors are not crowd-simulable out of the box (per-room walls, no doors, no navigation graph). The osm_clean toolkit now makes any imported floor sim-ready (synthesize doorways, build NRG, close scan gaps, place APs). That — plus the reusable importer, the counting/transfer reducer, and the atlas — is what persists.

Results

Goal 2 — flow & dwell (works). Three real floors, office agendas, tuned crowd; 244 agents, 22,482 WiFi device-traffic emissions, coupled CSI, all gate-passed. The atlas shows real paths threading real corridors and dwell hotspots where seated workers' devices emitted. Resolved: yes.

Goal 1 — counting fidelity (weak, floor-dependent). Device-free (fixed Tx→24 APs, body-perturbation) within-space R²: area1 0.64, area5 0.05, area6 0.01. Counting succeeds only where a ceiling AP happens to share line-of-sight with an occupied room. Device-based sensing is worse — only 3 of 7,128 seated-phone→AP links penetrate the walls. Resolved: yes, negative.

Goal 3 — cross-space transfer (fails). Train-on-A / test-on-B R² is uniformly negative (−3 to −67). A counter is geometry-specific and does not port between real floors — consistent with the thesis's "fingerprints don't transfer". Resolved: yes, negative.

The reframe (why this matters despite the null)

The occlusion is not a failure mode — it is the defining feature of indoor CSI sensing: an access point sees ~its own room. Three consequences reset the research question:

  1. CSI counting is per-room, not building-wide. Building occupancy = Σ per-room counts. The right question is a coverage/placement one: how many sensors, where, to count every room — the mall placement-oracle elevated to real multi-room buildings, where the walls define the decomposition.
  2. Passive device counting must be association-based, not ray-traced cross-room CSI: count the devices each AP serves (its room), not the CSI it can trace through walls.
  3. This is a direct argument for BLE-calibrated CSI (the thesis): a coarse room-assignment signal (BLE / device association) says which room a person is in; CSI counts within that room. Neither alone counts a walled building — which is exactly what this campaign showed. The failure of CSI-alone is the setup for the fusion win.

What it means / next

The genuinely valuable follow-on is a per-room counting + sensor-placement coverage study with BLE room-assignment fusion on these same real floors — turning the occlusion from the thing that defeated us into the structure the method exploits. Everything here is ray-traced simulation on real geometry: a hypothesis generator, and a strong motivation for the fusion architecture, until a real AX210/Pi5 walled floor (IP-106/IP-112) confirms the physics.

Figures: fig_device_counting.png (counting + transfer), the S3DIS floor + heatmap; interactive atlas as a published Artifact.

Criticism adversarial review

Written by the campaign-critic subagent against the brief's success criteria — read it as the counter-position to the synthesis above.

Criticism — c-public-spaces 01KY0RR3SS30H1P6NP9QXM434Y

🔴 The framing was naive (this is why the result feels thin)

"Count the whole building from one central Tx + scattered ceiling APs" is a strawman in a walled building — walls guarantee occlusion, so a null result was baked in. The finding "CSI counting fails building-wide behind walls" is physically near-obvious. Low insight-per-compute.

⚠️ What the campaign actually established (the useful kernel)

  • Occlusion is not a bug — it is the defining structure of indoor CSI: each AP senses ~its own room. → CSI counting is inherently per-room; building occupancy = sum of per-room counts.
  • Device-based (carried phone → AP) is even more occluded (3/7128 links) — passive device-association counting must use the device's own room AP, not ray-traced cross-room CSI.
  • Transfer fails hard (geometry-specific) — consistent with the thesis's fingerprint-non-transfer prior.

⚠️ Methodological caveats

  • bridge walkability polish is floor-dependent (over-dilated 2/3 floors) — atlas uses best-run-per-floor.
  • S3DIS is offices, not the transit/museum/mall public spaces originally intended (data availability).
  • Single seed per floor; sim-only; no hardware anchor.

Verdict

seal (as an honest limitation study + reframe). The deliverable value is the tooling + the atlas + the negative result that MOTIVATES the right next study (per-room counting + placement coverage + BLE room-assignment fusion) — not the counting numbers themselves.

Attached runs

Run Gate Purpose Replay
9WR2YGNF replay
PBPAGFXQ replay
KSEC1ZHC replay
F2AHPGX8 replay
0VXB8SAD replay
0GQK5N28 replay