EUVIMEDEUVIMED entdecken
Crossref

Advancing robust underwater acoustic target recognition through multitask learning and multi-gate mixture of experts

Yuan Xie iD, Jiawei Ren, Junfeng Li iD, Ji Xu

The Journal of the Acoustical Society of America · 01.07.2024 · 156 (1) · 244-255

DOI: 10.1121/10.0026481 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The Journal of the Acoustical Society of America, online erschienen am 09.07.2024. Verfasst von Yuan Xie und 3 weiteren Autor:innen. Beteiligte Einrichtung: Key Laboratory of Speech Acoustics and Content Understanding, Institute of Acoustics, Chinese Academy of Sciences, Beijing 100190, China. Die Arbeit wurde bisher 23-mal zitiert (Crossref) und nennt 68 Literaturangaben. Gefördert von: IOA Frontier Exploration Project, Youth Innovation Promotion Association of the Chinese Academy of Sciences, Chinese Academy of Sciences Strategic Leading Science and Technology Project.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Underwater acoustic target recognition has emerged as a prominent research area within the field of underwater acoustics. However, the current availability of authentic underwater acoustic signal recordings remains limited, which hinders data-driven acoustic recognition models from learning robust patterns of targets from a limited set of intricate underwater signals, thereby compromising their stability in practical applications. To overcome these limitations, this study proposes a recognition framework called M3 (multitask, multi-gate, multi-expert) to enhance the model's ability to capture robust patterns by making it aware of the inherent properties of targets. In this framework, an auxiliary task that focuses on target properties, such as estimating target size, is designed. The auxiliary task then shares parameters with the recognition task to realize multitask learning. This paradigm allows the model to concentrate on shared information across tasks and identify robust patterns of targets in a regularized manner, thus, enhancing the model's generalization ability. Moreover, M3 incorporates multi-expert and multi-gate mechanisms, allowing for the allocation of distinct parameter spaces to various underwater signals. This enables the model to process intricate signal patterns in a fine-grained and differentiated manner. To evaluate the effectiveness of M3, extensive experiments were implemented on the ShipsEar underwater ship-radiated noise dataset. The results substantiate that M3 has the ability to outperform the most advanced single-task recognition models, thereby achieving the state-of-the-art performance.

Autor:innen und Institutionen

  • Key Laboratory of Speech Acoustics and Content Understanding, Institute of Acoustics, Chinese Academy of Sciences, Beijing 100190, China.
  • University of Chinese Academy of Sciences, Beijing 100190, China.
  • State Key Laboratory of Acoustics, Institute of Acoustics, Chinese Academy of Sciences, Beijing 100190, China.

ORCID-Profile: Yuan Xie, Junfeng Li

Förderung

  • IOA Frontier Exploration Project
  • Youth Innovation Promotion Association of the Chinese Academy of Sciences
  • Chinese Academy of Sciences Strategic Leading Science and Technology Project

Zitieren

APA

Xie, Y., Ren, J., Li, J., & Xu, J. (2024). Advancing robust underwater acoustic target recognition through multitask learning and multi-gate mixture of experts. The Journal of the Acoustical Society of America, 156(1), 244-255. https://doi.org/10.1121/10.0026481

Vancouver

Xie Y, Ren J, Li J, Xu J. Advancing robust underwater acoustic target recognition through multitask learning and multi-gate mixture of experts. The Journal of the Acoustical Society of America. 2024;156(1):244-255. doi:10.1121/10.0026481

BibTeX
@article{Xie2024,
  author = {Xie, Yuan and Ren, Jiawei and Li, Junfeng and Xu, Ji},
  title = {Advancing robust underwater acoustic target recognition through multitask learning and multi-gate mixture of experts},
  journal = {The Journal of the Acoustical Society of America},
  year = {2024},
  volume = {156},
  number = {1},
  pages = {244--255},
  doi = {10.1121/10.0026481}
}
Kostenlos starten →