High resolution underwater imaging of complex objects using sparse sensor arrays

The problem of imaging a complex object submerged in shallow waters using a sparse surface sensor array is considered. To this effect, a hybrid signal processing method is developed, implemented, and assessed. This method is constructed by: (1) refining the Kirchhoff migration technique to incorporate a zoning of the sensors and an analysis of multiple reflections, and (2) combining this technique with a filtering scheme such as a direction of arrival estimation method or subspace filtering.;The proposed imaging method is assessed by numerical simulations. For this purpose, a high-fidelity computational acoustics framework is tailored to the solution in the time-domain of underwater acoustic scattering problems. Furthermore, an all-hexahedra automatic meshing procedure is developed for finite element computations in bounded domains around complex shapes. The overall computational framework is applied to generate high-quality pressure time-histories for various scatterers including a mockup submarine.;Using these pressure signals, the performance of the proposed signal processing method for imaging submerged objects with complex shapes using a sparse surface sensor array is assessed. As long as the noise to signal ratio is less than 10%, it is shown that this method has a strong potential for identifying underwater intruders. In particular, it produces images that are rich enough to allow surface reconstruction by geometrical means, or be considered as initial guesses for a nonlinear inversion method thereby accelerating its convergence to more detailed images.