Innovative millimetre-wave radar sensors now enable more accurate and multi-layered occupancy detection in smart homes, surpassing traditional motion sensors with real-time, device-based recognition.
Traditional motion sensors were designed to spot movement, not sustained occupancy. That is why a room can go dark while someone is still reading, working or watching television: passive infrared sensors only react when heat moves across their field of view. For smart homes that need lighting to remain on while a person is still, that limitation makes ordinary PIR hardware a poor fit, according to the XDA Developers article.
A more dependable approach is to use millimetre-wave radar, which emits radio waves and reads the reflected signal to detect tiny changes in position. Guides for the LD2410 sensor on ESP32-based Home Assistant setups say this kind of hardware can detect both moving and stationary people, making it better suited to real occupancy detection than PIR. Makerfabs also describes its LD2450 module as able to track up to three human targets at once, with a detection range of roughly 6 to 8 metres and a ±60-degree coverage angle.
That radar layer can be paired with Bluetooth Low Energy beacons to identify who is present, rather than only whether someone is in the room. The XDA piece suggests using ESP32 nodes or Home Assistant’s Bermuda triangulation tools to read phone or watch advertisements and infer proximity from RSSI values. In practice, that lets automations do more than switch lights on: they can load individual preferences, desk presets or room-specific scenes when a particular device is detected nearby.
For edge cases, enthusiasts also combine radar with physical and software signals. The article points to load sensors under chairs or mattresses to confirm a seated or resting state when radar may be obstructed, such as beneath a thick duvet. It also notes that desktop activity, smart plug power draw and active-session telemetry can reinforce occupancy decisions when a person is using a computer, television or games console.
The broader case for these systems is that they stay local, respond quickly and reduce reliance on cloud services. Everything Presence Lite, for example, is described by CNX Software as an ESP32-based 24GHz sensor that can track up to three targets while also acting as a Bluetooth proxy. Other Home Assistant-focused products, including Everything Presence One and Screek’s mmWave line, combine radar with additional sensors to support lighting, heating and energy control. The result is a multi-layered presence model that is more accurate than a single motion trigger and better aligned with how people actually use their homes.
Disclaimer: This content is intended for informational purposes only. Readers are advised to exercise their own judgement, conduct due diligence, or consult a qualified expert before acting on any information provided.
