Espectre is an edge machine learning framework and motion detection platform that uses Wi-Fi Channel State Information to identify human presence and movement. It functions as a sensing toolkit for ESP32 microcontrollers, enabling the detection of motion through walls without the use of cameras or wearables.
The project distinguishes itself by executing compact neural network classifiers and mathematical detection algorithms directly on the microcontroller. It utilizes a MicroPython runtime to allow for the prototyping and deployment of sensing logic and wireless signal processing algorithms without the need for manual compilation.
The framework covers a broad range of signal processing and integration capabilities, including subcarrier analysis, noise filtering via Hampel-Butterworth pipelines, and 3D object localization. It provides connectivity for smart home automation through Home Assistant and MQTT, while offering tools for raw data collection, UDP streaming for external analysis, and a web-based dashboard for monitoring device statistics.
Deployment is supported via a command line tool for flashing firmware across multiple ESP32 hardware variants.
