Home · Strong theory
Casing-Pattern Detector
It spots someone checking out your house days before a break-in.
The physics is solid; our own validation is scheduled.
How it works
Each brief visit near a fence line or side window shows up as a short presence at a spot where people do not normally linger. Once means nothing. The same quiet spot, short stays, several days in a row — that pattern is the classic sign of a burglar scouting, and it is exactly what the system looks for.
In the physics: Each loitering event is an NLOS presence detection with an RTI location estimate; casing produces a statistically improbable spatio-temporal recurrence — repeated short dwells at the same low-traffic perimeter cell — that is distinguishable from postal workers and gardeners by dwell location, duration, and time-of-day distribution.
See it work
This is real pipeline output for Home, replayed. The exact reading depends on your room.
Go deeper
How the field becomes an answer
How we prove it
Own-hardware longitudinal: 60-day deployment at 3 houses with scripted 'casing' sequences (3 x 2-min dwells over 4 days) injected against natural traffic, measure detection lead time and false alerts per month; ESP32-wall validates the per-event NLOS layer.
Who it is for
"We alert you to the burglary that's being planned, not just the one in progress." For neighborhood-watch platforms and insurance-linked security programs.
runtime Runs today in LatentField 2.0 as presence
The matching solution
Complete first-room estimateHome Presence & Intrusion2 nodes · ~$90 parts · 45 min Request kit, subscribe, or set up →Compatible DIY hardware options
45 min · intermediate
60 min · intermediate