Research article

Extraction method of weak underwater acoustic signal based on the combination of wavelet transform and empirical mode decomposition

Department of Intelligence and Automation, Taiyuan University, Taiyuan 030032, China

^* Corresponding author: bestxjl@qq.com

Received: 25 December 2020
Accepted: 3 March 2021

Abstract

When studying underwater acoustic exploration, tracking and positioning, the target signals collected by hydrophones are often submerged in strong intermittent noise and environmental noise. In this paper, an algorithm that combines empirical mode decomposition and wavelet transform is proposed to achieve the efficient extraction of target signals in the environment with strong noise. First the calibration of baseline drift is performed on the algorithm, and then it is decomposed into different intrinsic mode functions via empirical mode. The wavelet threshold processing is conducted according to the correlation coefficient of each mode component and the original signal, and finally the signals are reconstructed. The simulation and experiment results show that compared with the conventional empirical mode decomposition method and wavelet threshold method, when the signal-to-noise ratio is low and there exist high-frequency intermittent jamming and baseline drift, the combined algorithm can better extract the target signal, laying the foundation for direction-of-arrival estimation and target positioning in the next step.