lokalRAIN – Locally Distributed Rain Measurement with Acoustic Sensors in IoT Networks

Measuring precipitation is technically and economically challenging. Existing methods are laborious and costly, which limits their use. As a result, precipitation data is often only available for selected locations, which makes it difficult to forecast local rainfall accurately and to respond immediately to heavy rain events. There is therefore a need for innovative measurement methods that are cost effective and can be used ubiquitously.  

The "lokalRAIN" research project addresses this challenge by developing a novel rain measurement method. This method makes use of the structure-borne sound generated when precipitation hits various surfaces, such as photovoltaic modules, car roofs or roof windows. The resulting acoustic vibrations are recorded by vibration sensors and evaluated using machine and deep learning (ML/DL) methods to provide accurate precipitation data in real time. The aim of the project is to develop acoustic sensor nodes for rain detection, which can be used in a robust network to provide ubiquitous coverage through sensor data fusion. The project will develop the required ML/DL algorithms and a prototypical sensor node hardware, focusing on connectivity, low power consumption, low production costs and low maintenance requirements.  

Applications

Precise precipitation data is critical for several applications: Real-time rain monitoring serves to improve situation assessment and prioritization of actions during extreme weather and heavy rain events. Insurance companies need detailed weather data for improved risk assessment. Urban planners and disaster management authorities need accurate precipitation patterns to plan and monitor flood areas and drainage systems. High-resolution data allows farmers to optimize their irrigation strategies, improving efficiency and conserving resources.