SVMD coupled with dual-threshold criteria of correlation coefficient: A self-adaptive denoising method for ship-radiated noise signal

Abstract

Due to the complexity of the marine environment, ship-radiated noise signal (SRNS) often contains harsh marine background noise. Hence extracting the features of SRNS becomes very difficult. Aim to eliminate the interference of background noise, this paper proposes a self-adaptive denoising method for SRNS, called SVMD-Dual-CC-WST, which not only uses the successive variational mode decomposition (SVMD) as a theoretical basis but also combines with the dual-threshold screening of the correlation coefficient (Dual-CC) criterion and wavelet soft threshold (WST). Firstly, different original signals are decomposed into several intrinsic mode functions (IMFs) by SVMD. Then, the CC of the IMFs is calculated, and the IMFs are adaptively divided into three categories using the Dual-CC criterion, including signal IMFs, noisy IMFs, and noise IMFs. Finally, the noise IMFs are discarded, and the noisy IMFs are denoised with WST and then reconstructed with the signal IMFs to obtain the denoising signals. Simulated signals and SRNS experiments prove the effectiveness of the proposed method.