Abstract
The sensitivity of a Bartlett processor used as an objective function in a matched-field approach to bottom recognition in a shallow-water environment is investigated. The study addresses the varying response of the processor to changes in the values of geoacoustic parameters, in different frequencies and ranges from the source. It is shown that certain parameters are better estimated in low frequencies and shorter ranges. This observation led to the development of a two-stage algorithm which searches for certain sets of parameters in different frequencies and ranges. Thus, the complexity and dimensionality of the problem are reduced, improving the efficiency of a genetic algorithm search, both in terms of quality of the results and computational cost. The algorithm is applied to synthetic data of a benchmark exercise validating the conclusions of the study.

This publication has 14 references indexed in Scilit: