Acousto-optical effect signal denoising algorithm based on ICEEMDAN-grey correlation degree

In the optical-based measurement of the speed of sound in seawater, considering the influence of the measurement system and the media environment, the signal is embodied as a noise signal with acoustic markers in addition to the signal, especially when the SNR of the signal is low, which poses a great challenge to the subsequent time-of-flight analysis of ultrasound, thus seriously affecting the accuracy of the results of the speed of sound measurements, and it is necessary to suppress the noise. In this paper, we propose a noise reduction method based on Improved Complete Ensemble Empirical Mode Decomposition with Adative Noise (ICEEMDAN)-gray correlation. The noise-containing optical signal is decomposed into a series of IMF components and residuals using ICEEMDAN, and then the gray correlation between each IMF component and the original signal is used as the demarcation point for dividing it into noise-dominant and signal-dominant parts, and the IMF components and residuals in the signal-dominant part are reconstructed to get the noise-reduced signal at last. The relative performance of the algorithm is evaluated on experimental data, for low SNR 9dB acousto-optic signal after the algorithm the SNR is improved to 34.6861dB and compared with the existing methods, the combined data results show that the method is superior to the existing methods with higher SNR, NCC and lower MSE.