A big data framework for urban noise analysis and management in smart cities

Abstract

Environmental pollution monitoring is a major concern in the development of smart cities. Nowadays, urban noise is one of the most relevant pollutants, so many networks of acoustic sensors have been deployed to measure sound pressure levels at various locations. These acoustic sensors collect huge amounts of data, which can be helpful to manage noise events in urban planning. In this paper, a big data framework is proposed to properly analyse the considerably large amounts of noise monitoring data and obtain useful information for urban planning. A map and reduce approach is proposed to process the massive data captured from acoustic sensor networks, mobile phones and open data platforms. Using the map and reduce model, several statistical environmental acoustic parameters, including both temporal and spatial indices, can be calculated. As an example application, two algorithms are implemented to evaluate both day-evening-night equivalent levels (Lden) and percentile levels (Ln). An experimental case with data obtained from the Dublin open data platform shows the benefits of this framework for urban noise analysis and management.