Your WiFi Router Is Already a Sensor. RuView Turns It Into One.
You've got a WiFi router. It's been filling your home with radio waves this whole time. Every time you move, breathe, or sit still, you disturb those waves in ways that are measurable. Most of us just use WiFi for internet and never think about it again. RuView is an open-source project that takes those disturbances and turns them into something useful: presence detection, breathing rate, heart rate, and even body pose—all without cameras or wearables.
What It Does
RuView is a WiFi sensing platform that captures Channel State Information (CSI) from low-cost ESP32 sensors and turns it into spatial intelligence. The core idea is straightforward: radio waves bounce around your space, and when people move or even breathe, they change the signal in detectable ways. RuView reads those changes and makes sense of them.
What you get out of it: presence and occupancy detection through walls, contactless breathing and heart rate monitoring, activity recognition (walking, sitting, gestures, falls), sleep quality tracking with apnea screening, and RF fingerprinting that can identify rooms or detect moved furniture. There's also camera-free pose estimation—17 body keypoints derived from WiFi CSI.
The project integrates natively with Home Assistant, Apple Home, Google Home, Amazon Alexa, and Matter. A single --mqtt flag drops it into any Home Assistant install. It ships 21 entities per node—11 raw signals and 10 inferred semantic states like someone-sleeping, possible-distress, room-active, elderly-inactivity-anomaly, bathroom-occupied, and fall-risk-elevated. There are also three starter Home Assistant Blueprints included.
Under the hood, there's a built-in model workflow: record CSI, train models, load RVF files, and switch LoRA profiles. HOMECORE handles local automation with state, history, signed Wasm plugins, voice hooks, and HomeKit support. A unified RF world model lets you combine WiFi CSI with radar, UWB, and cellular sensing in one privacy-bounded scene model—though the README notes accuracy is still synthetic until real-data validation.
Why It's Cool
-
It works through walls, in the dark, without cameras. This isn't a gimmick—it's a fundamentally different approach to sensing. No lenses, no line-of-sight requirements, no wearables. Just physics. For anyone who's uncomfortable with cameras in bedrooms or bathrooms, this is a meaningful alternative.
-
The smart-home integration is unusually thorough. Most hobbyist sensing projects stop at a dashboard or a serial monitor. RuView treats Home Assistant, Apple Home, Google Home, Alexa, and Matter as first-class targets. The 21 entities per node aren't just raw numbers—they're semantic states like "possible-distress" and "elderly-inactivity-anomaly" that you can wire directly into automations.
-
The vital signs angle is genuinely interesting. Contactless breathing and heart rate monitoring while someone sleeps or sits still opens up use cases around elder care, sleep quality tracking, and apnea screening that would otherwise require dedicated medical devices or wearables.
-
It's honest about its limitations. The README explicitly says the unified RF world model's accuracy is "still synthetic until real-data validation." That kind of transparency is rare and worth noting.
-
The model workflow is built in. You don't need to leave the platform to record CSI data, train models, load RVF files, or switch LoRA profiles. That's a full pipeline, not just a signal reader.
-
Governed evidence is a thoughtful touch. Attaching privacy policy, uncertainty, provenance, and witness records to sensing events isn't something you see in most hobbyist projects. It suggests the authors are thinking about real-world deployment, not just demos.
How to Try It
The project lives at github.com/ruvnet/ruview. You'll need ESP32 sensors to capture CSI data—that's the hardware layer.
-
Clone the repository:
git clone https://github.com/ruvnet/ruview cd ruview -
For Home Assistant integration, run with the MQTT flag:
# Drop into any Home Assistant install with one --mqtt flag -
For Apple Home, Google Home, Alexa, or SmartThings, pair as a Matter Bridge.
-
Check the integration docs for your platform:
The README references HA-DISCO MQTT publisher for Home Assistant and a HAP-1.1 bridge for Apple Home. Siri, Google Assistant, and Alexa can voice presence and vitals by room with zero custom skills.
Final Thoughts
RuView is best suited for developers and tinkerers who already have some ESP32 hardware and are comfortable with Home Assistant or Matter integrations. It's not a plug-and-play consumer product—you'll need to source sensors, flash firmware, and work through the setup. But if you're interested in WiFi sensing, this is one of the more complete open-source implementations out there, especially given the breadth of smart-home integrations and the built-in model training workflow. The honest caveat about synthetic accuracy in the RF world model is worth keeping in mind, but the core CSI-based sensing is the real draw here. If you've got a spare ESP32 and a curiosity about what your WiFi can actually see, this is worth a weekend.