Hospitality · Strong theory

GhostRoom — hotel occupancy truth engine

Knows which hotel rooms are truly empty — from the corridor, through the wall.

Strong theory Strong theory live in runtime

The physics is solid; our own validation is scheduled.

A person in a room leaves two traces in the signal: the gentle rhythm of breathing and the ripple of movement. Both pass through a single wall. So housekeeping learns the truth — stayover, vacant, or someone lingering after checkout — with nothing visible installed in the room.

In the physics: Presence produces breathing-band CSI modulation (0.1-0.5Hz) and motion Doppler that penetrates a single drywall partition; occupant count changes aggregate multipath variance. Through-wall LOS/NLOS discrimination is already demonstrated (AUC 0.69) and improves with mesh diversity.
Preparing the recorded fieldLoading real LatentField pipeline output…

This is real pipeline output for Hospitality, replayed. The exact reading depends on your room.

Bring this into the real roomFloor & Bar Pulse3 sensing nodes · ~$130 parts · about 90 min
Go deeper

How the field becomes an answer

Corridor ESP32 mesh breathing-band Doppler + presence detector (class A) occupancy-count regressor per room compartment z-score cross-room transfer so one calibration generalizes across identical room types.

How we prove it

ESP32-wall dataset for through-wall presence; own pilot on one hotel floor comparing against PMS checkout records and housekeeping logs.

Who it is for

Housekeeping router + revenue-leak detector (unregistered guests, smoking-room parties) for the hotel operations director — pays for itself in cleaned-in-sequence labor savings.

runtime Runs today in LatentField 2.0 as occupancy_estimate

Complete first-room estimateFloor & Bar Pulse3 nodes · ~$130 parts · 90 min Request kit, subscribe, or set up →
Compatible DIY hardware options
Operator Single-Venue$280–380

120 min · advanced

← All Hospitality capabilities