A New Denoising Method for Underwater Acoustic Signal

Abstract

In recent years, the rapid development of marine science has put forward higher and higher requirements for the processing of ship-radiated noise signal. Ship-radiated noise is the noise signal generated by the vibration of various mechanical equipment or the movement of the hull and radiated into the sea when the ship is traveling. Ship-radiated noise signal contains a large number of time-varying, nonlinear and non-stationary components. The denoising processing of ship-radiated noise is the most critical part of underwater acoustic signal processing. In order to more effective reduce the noise of the ship-radiated noise signal, a new denoising method for underwater acoustic signal based on mutual information variational mode decomposition (MIVMD), multivariate multiscale dispersion entropy (mvMDE), and lift wavelet threshold (LWTD) and Savitzky Golay filter (S-G filter), named MIVMD-mvMDE-LWTD-SG, is proposed. Firstly, MIVMD is used to decompose the original signal into n subsignals. Secondly, the mvMDE value of each sub-signal is calculated, and the n sub-signals are divided into high-frequency components and low-frequency components according to the threshold. Then, S-G filter and LWTD method are used to reduce the noise of low-frequency components and high-frequency components respectively. Finally, the low-frequency components and high-frequency components after the denoising processing are reconstructed to obtain the denoising signal. In order to verify the effectiveness of the proposed method, the proposed method is used to reduce the noise of chaotic signal under different signalto-noise ratios (SNR), and compared with the EMD-mvMDE-LWTD and MIVMD-mvMDE-LWTD method. The results show that the proposed method can effective remove the noise in the chaotic signal, better distinguish the adjacent trajectories in the phase space, approximate the real chaotic attractor trajectory, and better retain the useful information in the chaotic signal. The proposed method is further applied to the actual ship-radiated noise signal, and the experimental analysis shows its effectiveness, which lays a solid foundation for further prediction and detection. INDEX TERMS Mutual information variational mode decomposition; multivariate multiscale dispersion entropy; lift wavelet threshold; chaotic signal; denoising

A new hybrid prediction model of cumulative COVID-19 confirmed data

“…Firstly, each IMF component obtained by VMD is defined as a signal with adjustable frequency and amplitude: where, is the kth modal component, and are instantaneous amplitude and phase respectively. The principle of VMD is as follows ( Yang et al, 2020 ): Hilbert transform is performed on , and unilateral spectrum of modal function is obtained by establishing analytical signal: Mix the analytical signal and center frequency of each IMF to realize the spectrum shift to the base band: Estimate the bandwidth of each IMF component. Constraint variational model: where, and represents the set of sub-signals and their center frequencies, K represents the total number of sub-signals, is the Dirac distribution.…”

Compressed Sensing Multiscale Sample Entropy Feature Extraction Method for Underwater Target Radiation Noise

“…However, when the denoising treatment is used, the radiated noise line spectrum cannot be completely recovered, limiting the implementation of the multiscale approach. In recent years, the compressed sensing method [19,20] has been frequently used in underwater acoustic signal denoising. Kim et al [21] investigated the tone signal produced by the mechanical component of the underwater target.…”

A Vibration Signal Denoising Method of Marine Atomic Gravimeter Based on Improved Variational Mode Decomposition

“…Hence, it could satisfactorily resist noise and reduce mode mixing. With its solid mathematic basis for the decomposition of nonstationary signals, VMD has been applied in earthquake time-frequency analysis, signal filter denoising, and ground vibration attenuation, which is a sufficient proof of its effectiveness and superiority in signal decomposition [22][23][24]. Nevertheless, number of decomposed modes K and penalty term α must be artificially set in the VMD of signals.…”