ruvnet/ruview

RuView is an open-source platform that turns ordinary WiFi signals into a sensing system capable of detecting people, tracking movement, estimating body poses, and measuring vital signs, all without cameras or wearables. The core insight is that WiFi radio waves already permeate every room, and when a person moves, breathes, or sits still, they disturb those waves in subtle, measurable ways. RuView captures and analyzes those disturbances. The technical mechanism relies on a concept called Channel State Information (CSI), which is detailed data about how a WiFi signal changes as it travels through a space. Low-cost ESP32 microcontroller chips (around $9 each) can capture this data when flashed with RuView's custom firmware. The captured signal data is then processed through digital signal filtering and neural network models to extract meaning: breathing rate is detected by filtering for slow oscillations (0.1, 0.5 Hz), heart rate by faster ones (0.8, 2.0 Hz), and body pose estimation maps WiFi disturbance patterns to 17 body keypoints using a technique derived from Carnegie Mellon University research called DensePose From WiFi. The system is designed to run entirely on edge hardware, meaning the ESP32 sensors and a small local computing device called a Cognitum Seed, with no cloud connectivity required. Multiple ESP32 nodes can form a mesh to improve spatial resolution. Real-world use cases include contactless sleep monitoring, fall detection for elderly care, occupancy counting for smart buildings, and security presence detection, anywhere cameras raise privacy concerns or are impractical. This is a hardware-software project requiring either ESP32-S3 microcontrollers or a research-grade WiFi network adapter, plus familiarity with embedded systems and signal processing. The tech stack is Rust for core processing code, with ESP32 firmware, Python for tooling, and Node.js for supporting scripts. A Docker image is provided for software-only evaluation with simulated data.