About this project
ESPectre is an open-source platform for private, local Wi-Fi motion sensing using Channel State Information (CSI) from a normal 2.4 GHz Wi-Fi 4 network. It processes sensing on supported ESP32 devices and reports a motion state and movement score without requiring cameras, microphones, radar hardware, wearables, or cloud connectivity.
The project provides multiple firmware paths: Native for standalone sensors, MQTT and Home Assistant MQTT Discovery; ESPHome for Home Assistant entities, provisioning and Device Builder updates; Matter for occupancy-sensor controllers, with compatibility still being validated; and Micro-ESPectre for lightweight MicroPython sensing with local read-only Direct HTTP monitoring. It also includes browser-based flashing, device settings and monitoring tools, plus a host CLI for setup, flashing and operator workflows.
For developers, ESPectre includes an embeddable C++ SDK, a Direct HTTP API, documentation for CSI acquisition, discovery, algorithms and architecture, and research assets such as CSI datasets, model weights, training/export/validation workflows, feature history and performance reports. Supported hardware listed in the README includes ESP32-C6, ESP32-C5, ESP32-C3, ESP32-S3, ESP32-S2 and classic ESP32.
ESPectre is positioned for room automations such as lights, displays, HVAC responses, alarms or notifications. The README cautions that it detects radio-environment changes, does not identify or count people, cannot prove a room is empty, and is not a replacement for safety-certified security, medical or emergency systems. First-party code is GPLv3, with separate commercial licensing available for eligible proprietary or closed-source firmware use.
Comments
0 Rating appears after 10 ratings
Sign in to join the discussion.