Ocean botton recognition from acoustic reflection data

In this paper we present an algorithm for the density and sound speed reconstruction of a two-layered bottom consisting of a fluid sediment layer over a fluid substrate of semi-infinite extend. The thickness of the sediment as well as the attenuation coefficients in the botton layer are also recovered. Measurements of the reflection coefficients of plane waves for eight source frequencies incident normally upon the water bottom interface are needed. The minimum number of frequencies needed for the algorithm is eight as it will be seen in the following sections, however if the reflection coefficient is known for more than eight frequencies the method makes use of the excess frequencies, improving the final solution in a postprocessing way. We first introduce our model, explain the method of solution and present the major steps of the algorithm. Then we will show results using synthetic data, for four cases which were tested, as well as results from a tank experiment.<>