Abstract
Recent advances in underwater sonar detection have highlighted the challenges of acquiring real-world sonar data due to experimental costs and environmental constraints. While deep learning-based methods have shown promise in sonar synthesis, existing approaches often lack fidelity in waveform generation. This manuscript proposes Multi-Band HiFi-GAN, a high-fidelity synthesis method that incorporates a multi-band processing module based on a Pseudo-QMF bank for spectrally-efficient sub-band decomposition, and Multi-Band STFT Discriminator (MBSD) that jointly models time-frequency structures and mitigates aliasing artifacts. Additionally, we design a multi-resolution STFT loss to enhance convergence. Evaluated on DeepShip dataset using Fréchet Audio Distance (FAD), our method demonstrates superior performance, achieving a lower FAD score than existing approaches. Augmenting real training data with synthesized samples improves target classification accuracy to 96.94% at a 100% mixing ratio. The work provides a practical data augmentation solution for underwater acoustic systems, improving synthesis fidelity.
Funding Information
  • Zhejiang Provincial Natural Science Foundation of China (LQN26F030024)
  • Central Government-guided local Science and Technology development (2025ZY01045)
  • Stable Supporting Fund of Acoustic Science and Technology Laboratory (JCKYS2024604SSJS00502)
  • National Natural Science Foundation of China (U20B2074)
  • National Natural Science Foundation of China (62471169)
  • Zhejiang Provincial Key Laboratory of Brain Computer Collaborative Intelligence Technology and Applications (2025E10015)