A frontier proposal with a validation plan, not an implemented detector.
Intended capability
Research whether the field could support this proposed outcome: Overnight per-stall restlessness index for horses: pawing, repeated lying/rising, rolling — the classic colic prodrome pattern — alerting the barn manager hours before morning rounds.
This describes the intended outcome. Readiness is research concept, evidence is class C, and a catalog mapping or recording is not proof of this outcome at a real site.
Solution blueprint
See the environment before installing it.
This exact kit is one of 191 first-class designs. It includes geometry, objects, nodes, wording, scenarios, installation, limitations, and catalog-bound readiness.
No rendered revision is available yet.
Bundled recording
A related Agriculture scene
This is one recorded vertical scenario. It is not separate validation of every capability in the catalog.
01 The physics
One horse in one stall is the shape the mmWave radar is built for: a single composite target in an aimed cone, whose motion and stillness energy rise and fall with the whole-body restlessness the colic prodrome is made of. Stall partitions attenuate the 24 GHz cone, so each stall's radar reads chiefly its own horse — the per-stall attribution the ambient field cannot promise. The CSI band shapes still carry the event identities: pawing is a rhythmic 1-2Hz limb band, lying/rising are high-energy posture transients (the proven fall/posture class), rolling is a sustained whole-body broadband burst. A quiet standing/sleeping horse is near-silent above the breathing band.
02 Shared processing path
Per-stall radar motion/stillness energy -> per-night restlessness index vs the horse's own baseline (the primary index) -> CSI Doppler band-shape templates classify the events behind a rise (paw rhythm / posture transition / roll) -> threshold alert with event replay.
This is a capability design path. Components may be shared with other catalog entries; it is not presented as a unique algorithm.
03 Validation plan
Own-hardware in a boarding barn with overnight video for labeling only; report event-class confusion and alert lead vs veterinarian-confirmed colic cases opportunistically. No public dataset — frontier build.
04 Commercial hypothesis
For the barn: 'The night watchman that never sleeps through pawing at 3am.' Plus tier, per instrumented stall: each watched stall carries its own mmWave radar, and only instrumented stalls are watched.