Determining the origin of impulsive noise events using wireless sound sensors

Environmental noise is a serious and growing pollution problem, with over 100 million people affected in Europe just by road traffic. Measuring and monitoring noise using a wireless sensor network is therefore getting more common. However, many scenarios consist of multiple sources of noise, with each source varying over time. That presents a challenge in determining which noise source is the cause of high noise levels, which is necessary to identify to reduce the problem. This work investigates how to identify the source of impulsive noise events using a pair of wireless noise sensors. One sensor is placed at a known noise source, and another sensor is placed at the noise receiver. Machine learning models receive data from the two sensors and estimate whether a given noise event originates from the known noise source or another source. In order to avoid privacy issues, the approach uses on-edge preprocessing that converts the sound into privacy compatible spectrograms and sound level representations. These different representations of sound can be stored, combined, and processed at a central server. The system was evaluated at a shooting range and explosives training facility, using data collected during noise emission testing. The combination of convolutional neural networks with cross-correlation achieved the best results (denoted bundled model). We created multiple bundled models with different stacking of delta-spectrograms. The best model detected 70.8\% of the impulsive noise events and correctly predicted 90.3\% of the noise events in the optimal trade-off between recall and precision score