No abstract

No abstract

This paper presents different neural network approaches for geoacoustic inversion. The basic idea of neural inversion is to approximate the inverse function from a set of behaviors, i.e. relations between acoustic fields and geoacoustic parameters. In this work, such methods have been applied in two different forms: a global approach which aims to estimate all parameters from all data, and a hierarchical approach in which the most sensitive parameters are estimated before the least sensitive. The methods are tested using synthetic data. Statistical results, as well as benchmark results show that such approaches are efficient and have similar performances.

Abstruct-In the summer of 1996, an experiment was conducted off the coast of Portugal to study the effects of internal tides on sound propagation. This experiment-called INTIMATE '96 (Internal Tide Investigation by Means of Acoustic Tomography Experiment)-has provided a great deal of insight about the variability of pulse transmission over space and time. In contrast to a common view of shallowwater propagation as complicated and unpredictable, we find a steady pattern of echoes. The echo-pattern stretches and shrinks in a systematic way with the tides and allows us to infer the components of the fwst few oceanographic modes. We also used the echo-pattern to track the source over a period of several days. During this period the isotherms in the ocean wavered by 20 m as a result of the tides, providing a challenge for modelbased tracking. We will discuss these acoustic results with emphasis on the source tracking.

No abstract