Multicomponent seismo-acoustic surface waves beamforming for undersea buried object detection

A beamforming method adapted to Stoneley-Scholte waves is presented, in the context of buried objects detection in the seabed. All the wave particularities are exploited in a wide-band multicomponent array processing: velocity, dispersion and polarization. These propagation features must be estimated. The incident wave is used for this purpose. Then a signal model is defined according to Stoneley-Scholte wave's propagation, in order to derived optimal and suboptimal receptors. The localization simulations show a good localization behavior for high Signal to Noise Ratios: the performances are close to the asymptotic estimation bounds. For high noise levels, the detection-performances depend on the noise spatial coherence. Detection is possible for signal to noise ratios as low as -20dB. Lastly, simulations with finite difference signals validate the model and demonstrate that four components are useful in the beamforming.