Abstract
Line spectrum is an important feature for the detection and classification of underwater targets. This letter presents a method for extracting the line spectrum submerged in underwater ambient noise through autoassociative neural networks (AANN). Compared with the traditional methods, the proposed method based on AANN can directly enhance the line spectrum from the raw time-domain noise data without relying on prior information and spectral features. Moreover, the proposed method can suppress the background noise while extracting the line spectrum. Both the numerical simulation and experimental data test results demonstrate that the proposed method provides a good ability to extract the line spectrum from the strong background noise.
Highlights
The radiated noise of underwater and surface targets contains many single-frequency components, i.e., of line spectrum;1 and the line spectrum of radiated noise carries a large number of target characteristic information, which is of great significance for detecting and classifying the targets.2Most of the existing methods of line spectrum extraction are based on the common spectral estimation method, such as detection of envelope modulation on noise (DEMON)3 or low-frequency analysis and recording (LOFAR).4 In addition, the time-frequency analysis methods, e.g., wavelet transformation5 and ensemble empirical mode decomposition (EEMD),6 are used for line spectrum extraction
We focus on the autoassociative neural networks (AANN), a type of artificial neural network, to extract line spectrum from underwater ambient noise
The signal-to-noise ratio (SNR) used in this letter is defined as SNR 1⁄4 10 log ðPs=PnÞ, where Ps and Pn represent the power spectral density (PSD) of the simulated signal and noise data, respectively
Summary
The radiated noise of underwater and surface targets contains many single-frequency components, i.e., of line spectrum; and the line spectrum of radiated noise carries a large number of target characteristic information, which is of great significance for detecting and classifying the targets.2Most of the existing methods of line spectrum extraction are based on the common spectral estimation method, such as detection of envelope modulation on noise (DEMON) or low-frequency analysis and recording (LOFAR). In addition, the time-frequency analysis methods, e.g., wavelet transformation and ensemble empirical mode decomposition (EEMD), are used for line spectrum extraction. The radiated noise of underwater and surface targets contains many single-frequency components, i.e., of line spectrum; and the line spectrum of radiated noise carries a large number of target characteristic information, which is of great significance for detecting and classifying the targets.. Most of the existing methods of line spectrum extraction are based on the common spectral estimation method, such as detection of envelope modulation on noise (DEMON) or low-frequency analysis and recording (LOFAR).. The time-frequency analysis methods, e.g., wavelet transformation and ensemble empirical mode decomposition (EEMD), are used for line spectrum extraction. The aforementioned methods utilize the spectral features or need prior information and artificial empirical adjustment. We focus on the autoassociative neural networks (AANN), a type of artificial neural network, to extract line spectrum from underwater ambient noise
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
