All capabilities
Residential Security Shared-family mapped

GaitKey

Resident-vs-Stranger Gait Fingerprint

Consented gait-similarity research; it cannot identify a resident or prove a stranger is present.

Readiness Shared-family mapped

Mapped to a shared offline runtime family and usable with a recording or compatible ESP32 CSI stream.

Evidence Strong theory Supported hypothesis

Published physics or adjacent results support the hypothesis; capability-specific product and site validation are still required.

Runtime family Gait signature

Repeatable movement-signature comparison, not biometric identification.

When no ESP32 is connected

No compatible ESP32 stream is connected. Use the bundled Home recording; it demonstrates the shared Gait signature runtime, not independent proof of this capability.

Intended capability

Pilot this intended outcome through the shared Gait signature family, then validate it against site-specific ground truth: The system learns each household member's walk within days and flags an unrecognized human gait inside the home — a 'someone is walking in your house and it isn't any of you' alert, without face recognition or tags.

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.

Shared-family mapped

No rendered revision is available yet.

Bundled recording

A related Home scene

This is one recorded vertical scenario. It is not separate validation of every capability in the catalog.

01 The physics

Walking produces a micro-Doppler signature: torso oscillation ~1-2 Hz plus limb harmonics up to ~10 Hz, whose cadence, stride-induced Doppler spread, and duty cycle are biometric-grade individual traits (established in WiFi gait-ID literature). Pets and robots occupy distinct micro-Doppler regions — machine-vs-human discrimination is already proven in LF2.

02 Shared processing path

Micro-Doppler feature extraction 0.1-40 Hz (stage 2) -> adaptive PCA builds per-resident gait manifolds with rolling baseline (stage 3) -> new gait = outside all enrolled MVE ellipsoids -> alert; cross-room z-score transfer applies enrollment house-wide.

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

SenseFi and CSI-Bench identity/HAR benchmarks for feature validity; own-hardware enrollment trial with 5 households x 4 people x 2 weeks, report stranger-detection AUC and per-resident confusion.

04 Commercial hypothesis

"Your house knows the difference between your teenager sneaking a snack and a stranger — and only wakes you for the stranger." For premium smart-alarm subscribers.