As part of the Swarmchestrate project, a forward-thinking Horizon Europe initiative, our InnoRenew CoE team is enhancing the urban noise monitoring system we operate in Koper. For this, we’ve developed custom sensor units ourselves, built around ESP32 microcontrollers for continuous sound pressure level (SPL) monitoring and Raspberry Pi 5s equipped with omnidirectional microphones to capture surrounding sounds when needed. Using Swarmchestrate’s decentralized approach, we’re upgrading these powerful sensors to perform initial data analysis directly on the device, rather than relying solely on the cloud.
This edge computing method improves data privacy, increases efficiency, and provides faster insights into noise pollution. We guarantee privacy through a specific process: when high noise levels trigger an audio recording via the microphone, an onboard AI immediately classifies the sound source. Crucially, the recording is then instantly deleted. It never leaves the device and is never stored, ensuring resident privacy is paramount. It’s an exciting step towards smarter, more responsive environmental monitoring for Koper, powered by cutting-edge European research and our unique, in-house developed technology.
