EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Sensor-Oriented Framework for Underwater Acoustic Signal Classification Using EMD–Wavelet Filtering and Bayesian-Optimized Random Forest

Sergii Babichev iD, Oleg Yarema iD, Yevheniy Khomenko iD, Denys Senchyshen iD, Bohdan Durnyak iD

Sensors · 28.08.2025 · 25 (17) · 5336

DOI: 10.3390/s25175336 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 28.08.2025. Verfasst von Sergii Babichev und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Physics, Kherson State University, 73008 Kherson, Ukraine. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 23 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Ship acoustic signal classification is essential for vessel identification, underwater navigation, and maritime security. Traditional methods struggle with the non-stationary nature and noise of ship acoustic signals, reducing classification accuracy. To address these challenges, we propose an automated pipeline that integrates Empirical Mode Decomposition (EMD), adaptive wavelet filtering, feature selection, and a Bayesian-optimized Random Forest classifier. The framework begins with EMD-based decomposition, where the most informative Intrinsic Mode Functions (IMFs) are selected using Signal-to-Noise Ratio (SNR) analysis. Wavelet filtering is applied to reduce noise, with optimal wavelet parameters determined via SNR and Stein’s Unbiased Risk Estimate (SURE) criteria. Features extracted from statistical, frequency domain (FFT), and time–frequency (wavelet) metrics are ranked, and the top 11 most important features are selected for classification. A Bayesian-optimized Random Forest classifier is trained using the extracted features, ensuring optimal hyperparameter selection and reducing computational complexity. The classification results are further enhanced using a majority voting strategy, improving the accuracy of the final object identification. The proposed approach demonstrates high accuracy, improved noise suppression, and robust classification performance. The methodology is scalable, computationally efficient, and suitable for real-time maritime applications.

Schlagwörter

Stichwörter der Autor:innen

bayesian optimizationempirical mode decompositionfeature extractionnon-stationary signal processingrandom forestship acoustic signal classificationunderwater acoustic sensorswavelet filtering

Autor:innen und Institutionen

  • Department of Physics, Kherson State University, 73008 Kherson, Ukraine.
  • Department of Informatics, Jan Evangelista Purkyně University in Ústí nad Labem, Pasteurova 3632/15, 400 96 Ústí nad Labem, Czech Republic.
  • Department of Digital Economy and Business Analytics, Ivan Franko National University, 79000 Lviv, Ukraine.
  • Department of Computer Technologies in Publishing and Printing Processes, Printing Art and Media Technologies Institute, Lviv Polytechnic National University, 79000 Lviv, Ukraine.

ORCID-Profile: Sergii Babichev, Oleg Yarema, Yevheniy Khomenko, Denys Senchyshen, Bohdan Durnyak

Zitieren

APA

Babichev, S., Yarema, O., Khomenko, Y., Senchyshen, D., & Durnyak, B. (2025). Sensor-Oriented Framework for Underwater Acoustic Signal Classification Using EMD–Wavelet Filtering and Bayesian-Optimized Random Forest. Sensors, 25(17), 5336. https://doi.org/10.3390/s25175336

Vancouver

Babichev S, Yarema O, Khomenko Y, Senchyshen D, Durnyak B. Sensor-Oriented Framework for Underwater Acoustic Signal Classification Using EMD–Wavelet Filtering and Bayesian-Optimized Random Forest. Sensors. 2025;25(17):5336. doi:10.3390/s25175336

BibTeX
@article{Babichev2025,
  author = {Babichev, Sergii and Yarema, Oleg and Khomenko, Yevheniy and Senchyshen, Denys and Durnyak, Bohdan},
  title = {Sensor-Oriented Framework for Underwater Acoustic Signal Classification Using EMD–Wavelet Filtering and Bayesian-Optimized Random Forest},
  journal = {Sensors},
  year = {2025},
  volume = {25},
  number = {17},
  pages = {5336},
  doi = {10.3390/s25175336}
}
Kostenlos starten →