Adaptive denoising method of refractive index signals in underwater flow fields

To enhance the effective target detection in complex underwater environments, the adaptive denoising methods suitable for underwater wake refractive index signals were investigated. The wake refractive index variation signals were obtained through the experiment simulating propeller wake under density stratification conditions, and the adaptive time-frequency analysis methods were employed for the purpose of denoising.The denoising effects of 3 methods,variational mode decomposition (VMD), complete ensemble empirical mode decomposition (CEEMD), and empirical wavelet transform (EWT),were compared. The results indicate that VMD and CEEMD are less effective when separating wake signals from background noise under low signal-to-noise ratio conditions. In contrast, EWT demonstrates superior performance in adaptively separating background noise under these conditions without frequent parameter adjustments. Therefore, it can be concluded that EWT is suitable for denoising wake signals in complex underwater environments, contributing to improved the accuracy and stability of signal processing.