Abstract
One of the main aspects affecting the quality of life of people living in urban and suburban areas is the continuous exposure to high road traffic noise (RTN) levels. Nowadays, thanks to Wireless Acoustic Sensor Networks (WASN) noise in Smart Cities has started to be automatically mapped. To obtain a reliable picture of the RTN, those anomalous noise events (ANE) unrelated to road traffic (sirens, horns, people, etc.) should be removed from the noise map computation by means of an Anomalous Noise Event Detector (ANED). In Hybrid WASNs, with master-slave architecture, ANED should be implemented in both high-capacity (Hi-Cap) and low-capacity (Lo-Cap) sensors, following the same principle to obtain consistent results. This work presents an ANED version to run in real-time on Controller-based Lo-Cap sensors of a hybrid WASN, discriminating RTN from ANE through their Mel-based spectral energy differences. The experiments, considering 9 h and 8 min of real-life acoustic data from both urban and suburban environments, show the feasibility of the proposal both in terms of computational load and in classification accuracy. Specifically, the ANED Lo-Cap requires around of the computational load of the ANED Hi-Cap, while classification accuracies are slightly lower (around 10%). However, preliminary analyses show that these results could be improved in around 4% in the future by means of considering optimal frequency selection.
Highlights
The results prove the viability of the two-class classification scheme to detect anomalous noise events (ANE), using a window of 30 ms, and Mel Frequency Cepstral Coefficients (MFCC) to parameterize the audio, and GMM as probabilistic classifier, outperforming the one-class classification (OCC) counterpart in both sampled urban and suburban environments
We have analyzed the viability of an algorithm to implement an Anomalous Noise Event Detector (ANED) Lo-Cap to run real-time in the Lo-Cap acoustic nodes of an hybrid Wireless Acoustic Sensor Networks (WASN), following the same principle used for implementing the original ANED version designed for the high-capacity nodes
The main conclusion of the analysis conducted in this research is that the ANED Lo-Cap proposal is viable, both in terms of computational load and classification accuracy, maintaining a consistent performance with respect to the obtained by the Hi-Cap counterpart
Summary
Until recently, noise measurements in cities have been conducted by professionals, who record and analyze the data in specific locations and time periods by using certified sound level meters. Noise maps are generated from these noise level measurements by means of the application of complex acoustic models after data post-processing. These maps should be updated and published every five years to fulfill the END requirements for agglomerations with more than. 100,000 inhabitants, major roads, major railways and airports [2]. This approach becomes difficult to scale when more measurements and/or locations are needed, besides the questionable
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
