Detection of acoustic events with application to environment monitoring

Abstract

The goal of this work is to present different detection techniques and its feasibility for detecting unknown acoustic signals with general applicability to different noise conditions. These conditions replicate those commonly found in real-world acoustic scenarios where information about the noise and signal characteristics is frequently lacking. For this purpose, different extensions of the energy detector and even new structures for improving the robustness in detection are considered and explained. Furthermore, three different research lines of application are presented in which the energy detector and its extensions are used to improve the localization accuracy and the classification rates of acoustic sounds.