Please login first
Characterization of a WASN-based urban acoustic dataset for the Dynamic Mapping of Road-Traffic Noise
* , , ,
1  GTM - Grup de recerca en Tecnologies Mèdia. La Salle - Universitat Ramon Llull.


Road traffic noise (RTN) is one of the main pollutants in urban and suburban areas, negatively affecting the quality of life of their inhabitants. In the context of the European LIFE DYNAMAP project, a noise mapping system has been developed to determine the acoustic impact of road infrastructures in real-time. The project has deployed two Wireless Acoustic Sensor Networks (WASN): one in the District 9 of Milan (urban area), and another in the A90 motorway of Rome (suburban area). The system should be able to identify and remove those anomalous noise events (ANE) unrelated to regular road traffic present in both areas (e.g., sirens, horns, speech, doors, etc.) since its goal is to monitor only RTN, following the European Noise Directive 2002/49/EC. To do so, an Anomalous Noise Event Detector (ANED) has been included in the dynamic noise monitoring system running in real-time in the low-cost acoustic sensors to avoid biasing the computation of the equivalent traffic noise levels. After deploying the WASN in both pilot areas, two acoustic datasets have been built to adapt the previous version of the ANED algorithm to run in real-operation conditions using the data collected through each of the 24 low-cost acoustic sensor networks. In this work, we describe the preliminary results of the analysis of the 154h WASN-based urban acoustic dataset from Milan in terms of the characteristics of both RTN and ANEs (e.g., number of occurrences, signal-to-noise ratio, duration, etc.), considering the origin of the data according to the location of the sensors of the network. The results show that both the number and variability of ANEs are greater in the dataset of the urban area compared to those collected in the Rome suburban area, which underlines the importance of a specific training of the ANED algorithm for each environment.

Keywords: dataset; wireless acoustic sensor networks; low-cost sensors; dynamic noise monitoring; road traffic noise; anomalous noise events; urban acoustic environment