Radar vs PIR Motion Detection: How to Choose and Test a Camera
A camera can notice movement, decide whether it matters, start recording and send an alert. Those are four different jobs. PIR and radar describe sensing methods; AI person detection describes how a system interprets activity. Understanding that distinction helps you choose a camera and avoid expecting one feature to solve every false alert.
This guide compares the technologies and shows how to test them at a doorway, side path or yard. It uses current UCOCARE listings as examples, rather than claiming that every model uses the same detector.
PIR, radar and video detection compared
On smaller screens, scroll the table sideways to see every column.
| Method | What it responds to | What to check at your location |
|---|---|---|
| PIR: passive infrared | Changes in infrared energy across its sensing zones, commonly caused by a person moving through the scene. | Walking direction, distance, obstructions and the temperature difference between the subject and surroundings. |
| Radar | Radio signals reflected from objects. The information available depends on the radar design and processing. | The actual detection area and how well the camera rejects movement outside the area you care about. |
| Video motion detection | Changes in the camera image. | Moving vegetation, shadows, headlights and changing illumination. |
| AI classification | Patterns used to classify a detected subject, such as a person or vehicle. | Which categories the exact model supports and whether the setting filters alerts, recordings or both. |
PIR does not need visible light to sense movement. It is also different from the infrared LEDs that illuminate a night image: a camera can have IR night vision without having a PIR motion sensor. Panasonic explains the infrared-sensing principle behind PIR; its component performance figures do not establish the detection range of a UCOCARE camera.
Radar likewise does not depend on a brightly lit video image. However, radar detection does not by itself tell you that a recorded face will be clear. The camera still needs suitable lighting, framing and recording settings. Axis describes this complementary role in its radar surveillance overview.
Where these features appear in the UCOCARE range
The A90 mini WiFi camera two-pack is listed with AI/PIR detection, 2.4GHz WiFi and a 160° camera view. That viewing angle describes the image, not a verified PIR detection angle. Use it to shortlist a compact WiFi setup, then test the approach routes that matter.
The A70 four-camera solar kit lists radar-assisted motion detection and AI classification alongside a base station. That makes it a different system choice, not simply an A90 with another sensor. Camera quantity, outdoor placement, power, storage and the hub connection also affect the decision. We do not use a claimed false-alert percentage as a promise for your property.
Choose by the path a visitor takes
A narrow entrance: start with the point where you need the first useful recording. Ask someone to approach normally, pause, turn away and walk across the view. A camera that only detects them once they are standing at the door may be too late for your purpose, regardless of its sensor label.
A garden beside a road: separate the useful path from passing traffic. Start with a sensible detection area and moderate sensitivity, where those controls are available. Check whether unwanted alerts come from the road, a moving branch or an actual visitor before changing the settings.
A camera behind a window: do not assume PIR will detect people through ordinary glass. Glass interferes with the infrared energy PIR needs. Radar performance through a barrier is also model- and material-dependent; a radar label is not permission to hide the detector behind any surface.
A repeatable detection test
- Mark the intended boundary. Use an existing paving line or another visible reference. Decide where you want the recording to begin.
- Walk three routes. Approach directly, cross from the side and leave the scene. Repeat at a normal walking pace, in daylight and after dark.
- Check the saved clip. Note the first visible moment, whether the person stays in frame and whether the useful end of the event is recorded.
- Check the phone separately. Compare the saved event with the received notification. A recorded event without a phone alert can point to notification or connectivity settings rather than a sensing failure.
- Change one setting. Adjust angle, sensitivity or a supported zone, then repeat the same route. Keep the change only if it improves the result without creating an unacceptable gap.
A small log is more useful than impressions: record the route, day/night condition, whether a clip exists, its starting point and whether the phone alerted. There is no need to treat a handful of trials as a formal accuracy percentage.
Does radar mean continuous recording?
No. Detection technology and recording mode are separate specifications. A radar-equipped camera can still use event-based recording. To compare what happens between events, see our AOV, motion-triggered and continuous recording guide.
Choose the system that captures your required event reliably in its real position. PIR, radar and AI can each be useful, but the most persuasive evidence is a saved clip from the route you actually need to monitor.
By UCOCARE. Product information checked on 17 September 2026 against the current product listings and sources linked in this guide. Illustrations are supplied product images; practical checks are suggested routines, not claimed hands-on test results.