AI-based approach to detect, classify and localize underwater events. Preprint “Underwater Acoustic Networks for Security Risk Assessment in Public Drinking Water Reservoirs” available on arXiv

The preprint of the paper “Underwater Acoustic Networks for Security Risk Assessment in Public Drinking Water Reservoirs” is available on https://arxiv.org/abs/2107.13977

Authors: Jörg Stork, Philip Wenzel, Severin Landwein, Maria-Elena Algorri, Martin Zaefferer, Wolfgang Kusch, Martin Staubach, Thomas Bartz-Beielstein, Hartmut Köhn, Hermann Dejager, Christian Wolf

We have built a novel system for the surveillance of drinking water reservoirs using underwater sensor networks. We implement an innovative AI-based approach to detect, classify and localize underwater events. In this paper, we describe the technology and cognitive AI architecture of the system based on one of the sensor networks, the hydrophone network. We discuss the challenges of installing and using the hydrophone network in a water reservoir where traffic, visitors, and variable water conditions create a complex, varying environment. Our AI solution uses an autoencoder for unsupervised learning of latent encodings for classification and anomaly detection, and time delay estimates for sound localization. Finally, we present the results of experiments carried out in a laboratory pool and the water reservoir and discuss the system’s potential.

https://arxiv.org/abs/2107.13977