Near-Field Geoacoustic Inversion Using Bottom Reflection Signals via Self-Attention Mechanism
Yuxuan Ma,Xiaobo Zhang,Fukun Jiang,Zhengrong Wei,Chenguang Liu
DOI: https://doi.org/10.1109/jstars.2024.3405202
IF: 4.715
2024-06-15
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Abstract:Geoacoustic inversion typically involves the collection of far-field underwater acoustic data to obtain seabed geoacoustic parameters using empirical formulas and matched field inversion (MFI) techniques. However, acoustic data propagated over long distances can introduce inevitable errors in inversion results, and traditional MFI techniques suffer from low computational efficiency. Although deep learning technologies have been applied to geoacoustic inversion, conventional deep neural network (DNN) models struggle to capture the long- and short-term dependencies in bottom reflection data, leading to suboptimal inversion accuracy. These issues present challenges in rapidly and accurately acquiring geoacoustic parameters over large areas. To address this, we propose a near-field bottom reflection signal collection method, collecting bottom reflection signals over a wide range of grazing angles by drifting. Utilizing the characteristics of near-field sound propagation, we constructed the bottom reflection coefficient sequence dataset using the wavenumber integration method. We then introduce a novel deep learning model, self-attention geoacoustics, based on multihead self-attention mechanisms, which improves inversion accuracy. In addition, we propose an adaptive-weight multitask learning training strategy, significantly enhancing the prediction accuracy of sound attenuation. Experimental results demonstrate that our method outperforms conventional geoacoustic inversion methods based on MFI and DNNs in terms of efficiency and accuracy, proving the superiority of our approach.
engineering, electrical & electronic,imaging science & photographic technology,remote sensing,geography, physical