A Novel Underwater Acoustic Target Identification Method Based on Spectral Characteristic Extraction via Modified Adaptive Chirp Mode Decomposition

Autor: Zipeng Li, Kunde Yang, Xingyue Zhou, Shunli Duan
Jazyk: angličtina
Rok vydání: 2023
Předmět:
Zdroj: Entropy, Vol 25, Iss 4, p 669 (2023)
Druh dokumentu: article
ISSN: 25040669
1099-4300
DOI: 10.3390/e25040669
Popis: As is well-known, ship-radiated noise (SN) signals, which contain a large number of ship operating characteristics and condition information, are widely used in ship recognition and classification. However, it is still a great challenge to extract weak operating characteristics from SN signals because of heavy noise and non-stationarity. Therefore, a new mono-component extraction method is proposed in this paper for taxonomic purposes. First, the non-local means algorithm (NLmeans) is proposed to denoise SN signals without destroying its time-frequency structure. Second, adaptive chirp mode decomposition (ACMD) is modified and applied on denoised signals to adaptively extract mono-component modes. Finally, sub-signals are selected based on spectral kurtosis (SK) and then analyzed for ship recognition and classification. A simulation experiment and two application cases are used to verify the effectiveness of the proposed method and the results show its outstanding performance.
Databáze: Directory of Open Access Journals
Externí odkaz:
Nepřihlášeným uživatelům se plný text nezobrazuje