Training a U-Net based on a random mode-coupling matrix model to recover acoustic interference striations
This work addresses the recovery of acoustic interference patterns in underwater acoustics, which is incremental as it applies an existing U-Net method to a new domain with a synthetic data generation approach.
The researchers tackled the problem of recovering acoustic interference striations (AISs) from distorted ones in range-dependent waveguides with nonlinear internal waves, using a U-Net trained on data generated by a random mode-coupling matrix model. The result showed that the U-Net successfully recovered AISs under various signal-to-noise ratios and conditions, though the model was not physically accurate.
A U-Net is trained to recover acoustic interference striations (AISs) from distorted ones. A random mode-coupling matrix model is introduced to generate a large number of training data quickly, which are used to train the U-Net. The performance of AIS recovery of the U-Net is tested in range-dependent waveguides with nonlinear internal waves (NLIWs). Although the random mode-coupling matrix model is not an accurate physical model, the test results show that the U-Net successfully recovers AISs under different signal-to-noise ratios (SNRs) and different amplitudes and widths of NLIWs for different shapes.