XDA’s write-up on 60GHz radar sensors made the case that a lot of smart home owners already know: PIR motion sensors miss half the point. They fire when you walk in, drop out when you sit still, and cannot tell where in the room you are. mmWave radar sees micro-movements down to breathing, and the newest 60GHz modules add per-zone position tracking so a light only turns on for the sofa side of the room.
The hardware is only half the story. To turn a raw radar feed into useful automations, you need the desktop-side software that ingests it, splits zones, debounces false positives, and hands it off to lights, HVAC, and audio. We tested 7 apps for the smart-home hub side of a 60GHz radar setup on Windows, macOS, and Linux, and rated each on how quickly you can go from a bare sensor to a working per-zone automation.
What to look for in a radar presence-sensing app
- ESPHome or Zigbee support for the popular 60GHz radar boards (LD2410, LD2412, HLK-LD2450, EP1, mmWave FP2, Aqara FP2).
- Zone editing UI, ideally a top-down room grid where you drag rectangles to define chair, sofa, and doorway zones.
- Fast debounce logic so the light does not flicker when you cross a zone boundary.
- Combined sensor fusion with door contacts, PIR, or Bluetooth for the edge cases radar still misses.
- Local processing. Radar data is intimate. It should never leave the LAN.
- Cross-platform hub option, because your desktop is not always on and a Raspberry Pi or NAS often becomes the actual host.
Quick comparison
| App | Best for | Free | Open source | Local-only option |
|---|---|---|---|---|
| Home Assistant | Full smart-home hub with best radar integrations | Yes | Yes | Yes |
| ESPHome | Flashing the radar board itself | Yes | Yes | Yes |
| Zigbee2MQTT | Zigbee-based radar sensors like Aqara FP2 | Yes | Yes | Yes |
| Frigate | Camera fusion when radar alone is not enough | Yes | Yes | Yes |
| Node-RED | Visual automation for radar zones | Yes | Yes | Yes |
| openHAB | Java-based hub alternative to Home Assistant | Yes | Yes | Yes |
| Room Assistant | Bluetooth presence to complement radar | Yes | Yes | Yes |
The 7 best radar presence-sensing apps
1. Home Assistant, best overall hub
Home Assistant is the desktop-hosted or Pi-hosted hub that most radar tutorials assume. The mmWave integrations for LD2410, LD2412, EP1, and Aqara FP2 are first-class, and the built-in Areas system pairs cleanly with per-zone occupancy binary sensors. The Blueprints library has ready-made “wait X seconds after last presence” scripts so lighting behaves well without writing YAML.
Where it falls short: The learning curve is real. Zone definitions still happen in code for some sensor models, not a drag-and-drop UI.
Pricing:
- Free: Full app, self-hosted.
- Paid: Home Assistant Cloud is a Nabu Casa subscription for remote access and voice.
Platforms: Windows, macOS, Linux (including Docker and dedicated OS images).
Download: Home Assistant
Bottom line: Start here unless you already run openHAB. Nothing else has the integration coverage.
2. ESPHome, best for the radar board itself
ESPHome is the firmware toolchain most 60GHz radar owners end up using because the popular LD2410, LD2412, and HLK-LD2450 modules pair with an ESP32 that ESPHome flashes over USB or OTA. It reads the raw radar frame, exposes distance and angle as sensors, and publishes them to Home Assistant, MQTT, or openHAB. The YAML config for a full radar setup is under 30 lines.
Where it falls short: Compilation happens on the host machine and takes minutes per firmware change. Not the most beginner-friendly first step.
Pricing:
- Free: Full app.
- Paid: Free.
Platforms: Windows, macOS, Linux, Docker.
Download: ESPHome
Bottom line: The only sensible way to turn a bare radar board into a networked sensor. Install alongside Home Assistant.
3. Zigbee2MQTT, best for Zigbee radar sensors
Zigbee2MQTT makes Zigbee-based radar sensors (Aqara FP1E, FP2, and various Tuya white-label modules) work without the vendor cloud. It runs as a small Node.js service on the desktop or Pi and republishes the sensor to MQTT for Home Assistant or openHAB to consume. The FP2’s per-zone occupancy comes through as separate binary sensors out of the box.
Where it falls short: Needs a Zigbee coordinator dongle. Firmware quirks on new radar sensors sometimes ship in the Herdsman release before the main package.
Pricing:
- Free: Full app.
- Paid: Free.
Platforms: Windows, macOS, Linux, Docker.
Download: Zigbee2MQTT
Bottom line: Best pick when you would rather buy a Zigbee radar sensor than flash your own with ESPHome.
4. Frigate, best for fusion with cameras
Frigate is a local NVR that runs object detection on RTSP camera streams and can fuse the results with a radar feed to eliminate the last-mile false positives. Radar sees a body but not who; the camera confirms it is a person, not the dog. Frigate hands the fused signal back to Home Assistant as a single occupancy sensor.
Where it falls short: Requires either a Coral USB accelerator, an integrated GPU, or a modern CPU to keep inference latency low. Not free of hardware needs.
Pricing:
- Free: Full app.
- Paid: Frigate+ subscription adds cloud-trained models for edge cases.
Platforms: Linux native, Docker on Windows and macOS.
Download: Frigate
Bottom line: The pick for the two hard rooms in every setup: entryway and kitchen.
5. Node-RED, best visual automation layer
Node-RED sits next to Home Assistant or openHAB and turns radar zones into flowchart-style automations. Drag a “presence in zone A” node, wire it to a “wait 30 seconds” node, wire that to a “living room light off” node. Debugging is visual too, which is a real advantage when a zone starts triggering at 3 a.m. and you need to see what fired.
Where it falls short: Overkill if your automations fit in Home Assistant Blueprints. Two systems to maintain, not one.
Pricing:
- Free: Full app.
- Paid: Free.
Platforms: Windows, macOS, Linux, Docker.
Download: Node-RED
Bottom line: Add this once your radar automations grow past a dozen rules.
6. openHAB, best Home Assistant alternative
openHAB is the Java-based smart-home hub for anyone who would rather not run Python and YAML. Radar integrations arrive later than in Home Assistant but the core mmWave sensors are covered, and openHAB’s Item and Rule model handles per-zone presence cleanly once you learn its syntax.
Where it falls short: Smaller integration surface than Home Assistant. The learning curve trades YAML for Xtend, which is not obviously better.
Pricing:
- Free: Full app.
- Paid: Free.
Platforms: Windows, macOS, Linux.
Download: openHAB
Bottom line: Pick this if you already run a JVM stack and want your smart home under the same tooling.
7. Room Assistant, best BLE complement
Room Assistant is a Node.js daemon that tracks Bluetooth Low Energy beacons across multiple Raspberry Pi or NUC nodes, one per room, so you can tell which room a phone or wearable is in. Pair it with a 60GHz radar and you get the two signals that cover each other’s weaknesses: radar knows a body is in the room, Room Assistant knows which body.
Where it falls short: Needs a small computer per room. iOS advertising rate limits mean beacon presence updates less often than Android.
Pricing:
- Free: Full app.
- Paid: Free.
Platforms: Linux (Raspberry Pi is the canonical host), Docker on Windows and macOS.
Download: Room Assistant docs
Bottom line: Add this if you want per-person automations, not just per-room occupancy.
How to pick the right one
If you want the simplest working setup, install Home Assistant on a Raspberry Pi and buy an Aqara FP2 or a ready-made LD2410 board so ESPHome flashing is skipped.
If you already run a hub, skip Home Assistant and use ESPHome for a DIY radar sensor, Zigbee2MQTT for a Zigbee one, and let your existing hub handle the automations.
If your problem is false positives from motion sensors in a kitchen or hallway, put Frigate in the mix and let the camera confirm the radar hit.
If your automations are already sprawling, run Node-RED alongside the hub. Flow debugging beats YAML archaeology.
FAQ
What is the difference between mmWave and 60GHz radar?
They are the same thing in practical terms. Consumer smart-home radar sensors marketed as “mmWave” operate at 24GHz or 60GHz; the 60GHz modules give finer per-zone resolution and are the newer generation.
Can radar presence sensors replace PIR entirely?
For living areas, yes. For hallways and short-visit rooms, pair radar with a door contact. Radar shines when a person is stationary, PIR wins on quick entry detection at very low latency.
Do 60GHz radar sensors work through walls?
Consumer 60GHz modules are designed for a single room. Anything strong enough to penetrate walls reliably would raise regulatory issues and privacy questions the current sensors do not.
Does radar detect pets?
The better sensors expose a “target class” that lets you filter cats and small dogs. Larger dogs still trigger presence, so pair the sensor with an automation that checks pet-tracker location if it matters.
Is a Raspberry Pi enough to host all this?
A Pi 4 or Pi 5 runs Home Assistant, ESPHome, and Zigbee2MQTT together comfortably. Add Frigate and you want a mini PC with an integrated GPU or a Coral accelerator.