Acoustic source localization and discrimination in urban environments

Collaborative localization and discrimination of acoustic sources is an important problem for monitoring urban environments. Acoustic source localization typically is performed using either signal-based approaches that rely on transmission of raw acoustic data and are not suitable for resource-constrained wireless sensor networks or feature-based methods that result in degraded accuracy, especially for multiple targets. In this paper, we present a feature-based localization and discrimination approach for multiple acoustic sources using wireless sensor networks that fuses beamforming and power spectral density data from each sensor. Our approach utilizes a graphical model for estimating the position of the sources as well as their fundamental and dominant harmonic frequencies. We present simulation and experimental results that show improvement in the localization accuracy and target discrimination. Our experimental results are obtained using motes equipped with microphone arrays and an onboard FPGA for computing the beamforming and the power spectral density.