Mapped to a shared offline runtime family and usable with a recording or compatible ESP32 CSI stream.
Intended capability
Evaluate the sensing evidence for “Gait Decay Early-Warning (Fall Forecasting, not Fall Detection)” in a consented study using staged events and independent reference observations. This is not a care decision, fall prediction, emergency response, or proof of resident safety.
This describes the intended outcome. Readiness is shared-family mapped, evidence is class B, 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 Elderly scene
This is one recorded vertical scenario. It is not separate validation of every capability in the catalog.
01 The physics
A walking torso produces a bulk Doppler shift proportional to velocity (~10-25 Hz at 2.4GHz for 0.6-1.5 m/s), while swinging limbs add periodic micro-Doppler sidebands at stride cadence. Slowing gait, shortened stride and increased double-support time compress and blur this micro-Doppler signature in physically predictable ways; sit-to-stand shows a characteristic vertical-velocity chirp whose duration lengthens with lower-limb weakness.
02 Shared processing path
Doppler/micro-Doppler feature extraction (0.5-40Hz) per walking bout -> LD2410 radial-distance track on instrumented corridors gives direct walking speed, with RTI segmenting bouts (no path-length normalization needed) -> per-resident rolling baseline in adaptive PCA manifold -> compartment Kalman on daily gait-speed/cadence estimates + CUSUM over weeks for slow decline detection.
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
OPERAnet gait/activity segments for micro-Doppler gait feature extraction; own-hardware longitudinal study in 10 assisted-living units against monthly Timed-Up-and-Go and 4m walk-test ground truth, with the corridor radar's distance-over-time track as a per-bout walking-speed reference; fall AUC 0.8 already proven anchors the acute end.
04 Commercial hypothesis
For assisted-living executive directors: 'We tell you WHO will fall NEXT MONTH, not who fell last night' — a per-resident fall-risk trend line with zero wearables and zero cameras; plus tier: a corridor-aimed mmWave radar reads walking speed directly.