Over the decades, urban areas have seen a constant increase in ambient noise, which among other things is a result of the ever-accelerating economic and technological progress determining our everyday lives. Ambient noise is, and has always been, perceived by people as unpleasant and disruptive. But the problem goes much deeper: Today we know from numerous scientific studies that being subjected to noise on a daily basis can have severe negative consequences on people’s health. Especially temporary noise (caused by, for example, construction works at the house next door or ruthless drivers revving up the engine of their car on the nearby intersection) plays an important, yet difficult to define role.
For the “StadtLärm” (CityNoise) project, a consortium of partners has teamed up to develop a system for capturing, displaying, and predicting ambient noise occurring in urban areas. The system will be implemented in the form of a robust, cost-efficient sensor platform that can be used for large-scale, high-resolution noise measurement. The data gained from the measurements will be visualized by a 3-D representation and used for designing and testing noise-space models. Capturing data of high resolution with regard to both time and space will allow predicting future noise scenarios on the basis of past noise events. The system will have self-learning capabilities, which means that the models and services incorporated in the system become more and more accurate as the number and frequency of measurements increases. Users of the system would primarily be public authorities looking for new ways of analyzing and assessing ambient noise, particularly with regard to discrete noise sources and events.