Exploitable results
The results of this SITAR subtask is computer software for modelling of acoustic scattering from objects submerged in seawater, in particular, objects on the seafloor or buried under it. The software contains modules for forward modelling of such scattering as well as a module for identification of the scattering object by searching values of parameters such as size, shape, orientation or acoustic contrast, such that the model-predicted scattered field matches the experimentally observed data. The parameter search is done by either a differential evolution algorithm (DE) or a genetic algorithm (GA) for non-linear global optimisation, combined with local optimisation with the downhill simplex method.
The software is designed for modelling and analysis of scattering data collected with the emitter and the receivers in bi-static configurations. In the SITAR project, a configuration with a ROV-mounted parametric emitter and a vertical array of receiver hydrophones was used, however the software permits more general experimental configurations. The software contains two independent forward scattering models. The first uses an accurate full-field boundary-integral equation (BIE) method for computing the acoustic field scattered by a 3D object in a layered fluid-solid medium. The second is a fast scattering model based on acoustic ray tracing and kirchhoff's scattering approximation. The fast scattering model is used in the object identification module, while the BIE module is used for assessment of modelling accuracy. The software runs on single-host Linux systems as well as on Linux clusters.
The 3-D synthetic aperture sonar processing (SW) that we are developing at Ecole navale (+ARMINES) is meant to provide maps of the seafloor by using two parallel arrays (interferometry). The processing will be applied to both low frequency and high frequency ranges of the parametric sonar allowing an interface description in the high frequency (HF) range and objects localization in the low frequency (LF) one. The very important issue is, thanks to parametric arrays, both HF and LF maps will be obtained simultaneously and geographically co-located. In order to ensure having the same resolution in both frequency ranges, conventional synthetic aperture processing (SAS) will be applied to raw signal issued from both ranges. Interferometry will be achieved using wideband algorithms (time domain) and both LF and HF maps will be compared for pointing out buried targets. Analyses of sea data are on the way. Processing will also be applied to complementary data obtained during the system calibration in a large basin. The results achieved have been disseminated through several publications.
A set of software routines applicable to the compression, filtering, segmentation and feature extraction from 3-D acoustic scattering images has been implemented. The sw allows discriminating objects resting on or buried in the seabed from the seabed background.
A GIS-based data integration/presentation software for risk assessment of toxic seabed dumpsite. The sw can accept several classes of data, form environmental to biotoxicological to geographical, and display them accordingly to appropriate user-specified characteristics. All data are geo-referenced. Correlation among measurements, either in space and in time it is also possible.
Geoacoustic inversion of monostatic data from the main sea trial of the SITAR project has been carried out by NTNU. The method used, named FARIM - Frequency Analysis based Roughness and Impedance estimation Method, estimates the roughness and impedance of the sea bottom from calibrated sonar data. In presence of roughness at the bottom, energy is scattered incoherently away from the receiver. A closed form expression for the coherent part of return with RMS roughness height as parameter is obtained from Kirchhoff scattering theory. The underlying assumption is that roughness is a stochastic variable with a Gaussian probability density. The observed shift in the received signal spectrum, represented by the centre of gravity frequency, towards lower frequencies is then compared to the theoretical shift and an estimate for roughness is obtained. Calculating the bottom impedance simply from the ratio of received to transmitted energy generally results in an underestimation of impedance because of incoherent scattering. The estimated roughness is used to calculate an improved estimate of the bottom impedance.
The Multiple Aspect Scattering techniques that we have developed at the University of Bath are concerned with high-frequency acoustic scattering measurements in a multistatic set-up (hydrophones physically decoupled from the sonar transmitter). This is based on the design and conduct of experiments in a medium-sized basin with real seabeds representative of European environments, and using several types of scaled-down targets. These targets were proud, flush buried and half-buried. The influence of the seabeds, and the combination of target(s) and seabed(s), were investigated for a large range of incidence angles, scattering angles, and bistatic angles.
These systematic measurements have been used to prepare sea trials (in conjunction with our SITAR partners). They are the first complete dataset acquired at this range of frequencies, for such a wide range of targets and settings. The experimental methodology developed can be used as an example for similar studies.
Appropriate software techniques, based on advanced deconvolution software, have been developed to reconstruct the positions of the targets and their scattering characteristics (from the surface of the targets, and sometimes from the inside of the targets as well). Analyses of the tank experiment data and of the sea trials data are currently in progress. The results achieved so far are currently disseminated through S&T publications. More information can be obtained from the work package leader at the University of Bath.
Parametric sonars for sub-bottom profiling is being used to achieve higher spatial resolution than conventional sub-bottom profilers. This technology has now been developed further into a parametric side scan sonar (PSSS) system where high frequency primary signals are used for getting a conventional side scan image of the seabed and the parametrically generated difference frequency (low frequency) is used for penetration into the sediments in the same area. Two receiver arrays are used in order to perform interferometric processing for depth information.
The system operated at primary frequencies around 100 kHz and secondary frequencies from 4kHz to 20kHz. The transmitted signature can be either a Ricker wavelet, Chirp (FM) wavelet or a CW signal. The received signals will be processed by software developed by other partners in the SITAR project. The major software package is a 3-D synthetic aperture sonar-processing package.
The main purpose of PSSS is to improve the efficiency and reliability in surveying and detection of buried or partly buried objects such as waste, sources of pollution, mines etc.
Searching for OpenAIRE data...
There was an error trying to search data from OpenAIRE
No results available