Estimation of sediment properties using frequency domain identification and marine acoustics
Vandenplas, S.; Temsamani, A.B.; Van Biesen, L. (2001). Estimation of sediment properties using frequency domain identification and marine acoustics, in: IEEE OCEANS, 2001. MTS/IEEE Conference and Exhibition. Oceans (New York), 1-4: pp. 697-706 In: IEEE (2001). OCEANS, 2001. MTS/IEEE Conference and Exhibition. Oceans (New York), 1-4. IEEE: Honolulu, HI , USA . ISBN 0-933957-28-9. , more In: Oceans (New York). IEEE: New York. ISSN 0197-7385, more |  |
| Available in | Authors | | Document type: Conference paper
|
| Abstract | In the field of underwater acoustics, the importance of the characterization of the seafloor is well known and of interest for scientific, economical and military applications [1]. An acoustic method would reduce the present needs for sediment sampling, while producing continuous profiles of sediment properties. However, for studying the functional relations between acoustical and geo-physical properties, an accurate determination of the acoustical parameters is necessary first.
In this paper, a global system identification approach Is used for the experimental validation of wave propagation models through sediments and for the determination of its acoustical parameters. In order to achieve this goal, laboratory experiments on calibrated degassed sediments with broadband Panametrics piston transducers (300kHz-700kHz) are carried out. For this, two configurations of acoustic measurements are used: a water-Plexiglas-Sand-Plexiglas-water and a water-sediment-Plexiglas configuration.
The Maximum Likelihood Estimator is applied in the frequency domain to retrieve the sediment parameters. The propagation of multiples in the sediment, and the calibration method, which is incorporated in the global system identification approach, are important assets that are not exploited when carrying out direct measurements of dispersion or absorption with transducers buried in the sediment.
Hereby the wave propagation phenomena are represented as Single Input Single Output (SISO) transfer systems and modeled parametrically and with several types of propagation models (viscoelastic models, porous models, Buckingham's model). A comparison between the estimated parameters is carried out. Furthermore, it is demonstrated that the method can be extended to Multiple Input Multiple Output (MIMO) systems, where now several transfer functions are estimated together parametrically. It Is shown how the method can be used In real life experiments. |
|
