Glossary
The words we use, defined.
Short, plain definitions of the terms behind camera-free sensing in buildings: what fixture-native means, what mmWave radar does and does not capture, why presence is different from identity, and what a long lie is. Each term has its own anchor, so link to it.
- Fixture-native sensing
- Sensing built into the electrical fixtures a room already has to have, such as the light switch, the power outlet and the HVAC vent, rather than added as a separate device, camera or wearable. Because the fixture has to be there anyway, coverage is room by room, a retrofit is a swap, and nothing new is pointed at anyone. ALLIE by Arqaios is a fixture-native sensing platform.
- How the ALLIE platform works
- Ambient intelligence
- A building environment that senses what is happening in it and responds on its own, without asking the people inside to wear, carry, charge or operate anything. In ALLIE, ambient intelligence means the room notices a fall, an intruder or an empty space and acts: alerting staff, lighting a path or setting back the HVAC.
- ALLIE by Arqaios
- Presence versus identity
- The distinction between knowing that a person is in a room and how they are moving (presence) and knowing who they are or what they look like (identity). mmWave radar resolves presence, position, posture and motion as a point cloud and cannot capture a face, so it delivers safety and occupancy awareness without identity. Arqaios's privacy architecture is built on that distinction.
- Privacy and security
- mmWave radar
- Millimeter-wave radar: a short-range radio sensor that transmits at very high frequencies (ALLIE uses the 60 GHz band) and measures the reflections from people and objects to resolve distance, position, motion and even breathing. Its wide bandwidth gives resolution of a few centimeters, it works in the dark and through the whole room, and it produces a point cloud rather than an image.
- mmWave sensing in ALLIE
- Point cloud
- The output of a mmWave radar: a set of points in space, each with a position and a velocity, that together show where people and objects are and how they are moving. A person appears as a cluster of points with no face, skin or clothing, which is why radar can detect a fall without seeing anyone.
- Edge AI (on-device processing)
- Running the machine-learning models that interpret sensor data inside the device itself, in the room, instead of streaming raw data to a cloud service. In ALLIE, presence, activity, fall and distress classification happen in the fixture, so what leaves the room is an event rather than a recording.
- Edge AI in the fixture
- Event-based monitoring
- Monitoring that transmits and stores structured events, such as 'fall detected in bathroom at 02:14' or 'room occupied', rather than continuous video, audio or sensor streams. Event-based systems can support incident review and escalation without footage to store, leak or subpoena.
- Data principles
- Long lie (time on the floor)
- The time an older adult spends on the floor after a fall before help arrives. Lying on the floor for an hour or more is strongly associated with serious injury, hospital admission and moves into long-term care, and research in people over 90 found most call-alarm owners did not use their pendant after falling alone. Automatic fall detection exists to shorten the long lie.
- Fall prevention and detection
- Fall-risk patterning
- Using changes in a person's everyday movement, such as slower transfers, an unsteady gait, more night-time bathroom trips or a changed routine, as early warning of rising fall risk. ALLIE baselines each room's normal and surfaces these changes to care teams before an incident rather than after.
- Prevention is the point
- Occupancy-based setback
- Relaxing a room's heating, cooling, lighting and plug loads when the room is empty and restoring comfort before people return, driven by measured occupancy rather than a schedule, a keycard or a door contact. ENERGY STAR reports that hotel rooms sit unoccupied about 12 hours a day and that guest-room energy management can cut energy costs by 35 to 45 percent.
- Energy optimization
- Load shedding
- Automatically cutting power to non-critical loads, such as idle plug loads and empty-room HVAC, so that limited power goes to what matters, whether during a demand peak or on backup power in an outage. ALLIE outlets shed non-critical loads to extend generator and battery runtime, a behavior covered by U.S. Patent No. 12,282,307.
- Issued patents
- Standby waste (vampire load)
- Electricity drawn by devices that are switched off or idle but still plugged in, such as chargers, televisions and appliances in standby. ALLIE outlets identify always-on loads and cut them when the room is empty, restoring power instantly when someone returns.
- Energy capabilities
- Acoustic event sensing
- Recognizing the sound signature of a specific event, such as an impact, a call for help, breaking glass or a smoke-alarm chirp, on the device, without recording or transmitting audio. It is distinct from voice assistants and audio monitoring, which capture speech.
- Sensing and intelligence
- Duty of care (lodging)
- The responsibility a senior living community or hotel has for the safety of the people sleeping under its roof. In practice it means knowing when a resident or guest behind a closed door needs help, and getting help there quickly, without surveillance of everyone else.
- Hotels and extended stay
Next step
Questions about a term?
Ask the people who use them every day. A thirty-minute walkthrough covers sensing, privacy and what changes in your building.