Introduction to mmwave Sensors - Home Assistant Guide

mmWave Sensors: What They Are & How They Work With Home Assistant

mmWave sensors are a newer type of presence and motion sensor gaining popularity in smart homes - especially among Home Assistant users. Unlike traditional infrared (IR) sensors, mmWave uses millimeter‑wave radio frequencies to detect movement and even subtle presence, like someone sitting still.

What Is mmWave?

mmWave (millimeter wave) is a radio frequency range (30–300 GHz) corresponding to wavelengths of 1–10 mm. Smart‑home mmWave sensors emit radio waves that bounce off people or objects; by analyzing the returned signals, they detect movement, presence, and can estimate distance.

mmWave vs IR Motion Sensors

How They Detect: IR (also called PIR) sensors look for changes in heat (like your body moving). mmWave sensors send out invisible radio waves and notice when something moves by how those waves bounce back.
Sensitivity: mmWave can detect even small motions (like breathing); IR requires noticeable heat changes. IR sensors work well for things like detecting when someone walks into a room (to turn on lights etc.), but if someone is relatively still - like sleeping or sitting on the couch - they typically will not detect their presence. This is where mmwave comes in.
Through Objects: mmWave can sometimes "see" through thin walls or glass. IR cannot.
False Positives: mmWave may trigger from airflow, pets, or mechanical motion, while IR mostly triggers from heat sources.
Detection Speed: IR sensors usually notice people entering a room faster than mmWave sensors. mmWave is best at sensing when someone is staying still, while IR gives a quicker reaction for sudden movement or entry. So IR is a good fit for turning on lights when someone enters a room, and mmwave is good for knowing when the room is really no longer occupied and turning off the lights.

Common Use Cases

Accurate room occupancy - lights/heaters/fans stay on even when stationary.
Bedrooms, sitting rooms, bathrooms - avoid turning off lights when someone's stationary.
Security - detect lingering presence, even without movement.
Sleep tracking - some sensors gauge breathing or body motion.

Setting Up mmWave Sensors in Home Assistant

Different sensors use different connection types:

Zigbee: Pair with a Zigbee coordinator (e.g., Sonoff/HA SkyConnect).
Wi‑Fi: Often ESPHome‑based via USB or MQTT.
Bluetooth: Paired via BLE proxy or integrated radios.

Each protocol and device has unique pairing instructions - always refer to the vendor manual and HA integration docs.

Popular mmWave Sensors with Home Assistant

EP Lite (Everything Smart Home, ~€35–40, Wi‑Fi + Bluetooth):
Easy to set up and affordable. Supports multiple detection zones (up to 4), so you can map out areas of a room (like doorways or beds) and even exclude spots that cause false triggers (like fans). Zone setup can be done via Home Assistant.
Everything Presence One – EP1 (Everything Smart Home, ~€55–75, Wi‑Fi + Bluetooth):
More advanced hardware features, including person counting and environmental sensing (temperature and humidity). As of the latest beta firmware (1.23b), EP1 now also supports up to 4 "distance zones", letting you focus detection on specific areas by distance from the sensor - great for excluding unwanted triggers (like a fan). This feature is still considered beta but suggests that zone support could soon be rolled out to EP1 as well.
FP2 (Aqara, ~€70–95, Wi‑Fi + Bluetooth):
Offers advanced zone mapping (up to 30 zones in a single room), fall detection, person counting, AI-powered sleep tracking, and experimental AI person/pet filtering features.
FP1E (Aqara, ~€50–65, Zigbee):
Improved reliability and detection over the FP1, flexible local zone setup (not as detailed as the FP2).
SNZB‑06P (Sonoff, ~€16–22, Zigbee):
Budget-friendly and simple to integrate. Provides basic presence detection, making it ideal for small spaces or straightforward automations.

Note: Some Echo devices include mmWave presence detection, but their data is typically not exposed to Home Assistant - though some workarounds exist.

Common Issues & Pitfalls (Expanded)

While mmWave presence sensors are highly sensitive and versatile, this same sensitivity introduces some unique challenges and learning curves for smart home users:

False Triggers from Airflow & Moving Objects: mmWave sensors can pick up movement from fans, HVAC vents, waving curtains, robot vacuums, and even swinging pendant lights. Anything that moves - even slowly - within the sensor's detection field can generate a "presence" event.
Wall Bleed & Reflections: mmWave signals can sometimes penetrate thin walls or reflect off glass and metal, resulting in "ghost" detections in neighboring rooms or odd corners. Proper sensor placement - angled away from walls and windows - can reduce these effects.
Radio Interference: Placing sensors too close to large metal objects, Wi-Fi routers, or speakers can reduce their effectiveness. It's best to install mmWave sensors in open areas free of electronic or metallic clutter.
Complex Configuration: Many advanced sensors allow for defining detection "zones," ignore zones, or exclusion areas. Configuring these properly can help prevent false triggers from known sources like ceiling fans, pets, or doorways.
Feature Support Varies: Not every feature (like ignore zones, multi-zone detection, or AI filtering) is exposed through every Home Assistant integration. Some sensors require setup or mapping in a vendor app before zones and other features are available in Home Assistant.
Learning Curve: Sensitivity and presence timers often need tweaking. Expect some trial and error to achieve optimal reliability for your room and lifestyle.

Pets & mmWave Sensors (Expanded)

mmWave sensors do not differentiate between humans and animals - any sufficiently large or moving body, including pets, can trigger a presence event. For pet owners, this can be both a blessing (for whole-house security) and a headache (for routine automation). Here's how to minimize false triggers from pets:

Height & Angle Placement: Mount the sensor higher on the wall or ceiling and angle it away from common pet travel paths. Most pets (especially cats and small dogs) stay close to the ground, so a well-placed sensor can "look over" them while still detecting people.
Zone Exclusion/Ignore Zones: Many modern mmWave sensors (like the Aqara FP2 and Everything Presence Lite/EP1) support defining "ignore zones" or "interference areas" within their detection map. By marking where pets sleep or walk, you can tell the sensor to ignore movement in those specific regions.
Automation Logic: Use Home Assistant to combine mmWave with other sensors (like contact sensors on doors, pressure mats, or even traditional PIR sensors) to confirm true occupancy. For example, you might require both mmWave presence and a door opening to consider a room "occupied."
Schedule-Based Filtering: If your pets are active mostly during certain hours, schedule automations to be less sensitive at those times or only consider presence if another sensor is also triggered.
AI-Based Pet Filtering: The Aqara FP2 has added "AI person detection" that attempts to distinguish between pets and humans. However, community feedback suggests results are mixed - so it might be good to not rely solely on this feature for critical automations.
No True Automatic Pet Filtering (Yet): As of now, there are no consumer mmWave sensors that truly and reliably filter out pets in all conditions. This is likely to improve over time but at present, smart placement and ignore zones are your best bet.