Localização de fontes sonoras submarinas utilizando time-reversal mirror e o modelo de raios

Abstract

Underwater Source localization presents complex challenges related to sound propagation in this environment. This study focuses on passive localization techniques based on time reversal mirrors (TRM), investigating the method in conjunction with an underwater acoustic model. This dissertation explores the sound propagation models and signal processing methods necessary for implementing passive localization, rooted in the principle of reciprocity of the sound field and ray theory. The expressions of TRM were presented, and passive localization algorithms are established for both a single sensor and a vertical array of sensors. This study reveals that the correspondence between simulated and real acoustic channel is critical regarding the passive localization performance of TRM. Simulation results demonstrate that the error between the actual and estimated locations is less than 1m for a source at a distance of 500m, even in the presence of noise, although it may be sensitive to environmental uncertainties, especially variations of the sound speed in water. Furthermore, it was noted that increasing the number of sensors used in TRM processing resulted in better search performance. The result of experimental data with a vertical array reinforces the potential of TRM but points to the importance of vertical coverage of the water column for reducing ambiguity zones and improving accuracy of the proposed method.