AUTHOR=Zhao Danning , Lei Yu , Xu Jinsong , Cai Hongbing 

TITLE=A comparative study of four types of multi-scale entropies in feature extraction of underwater acoustic signals for potential GNSS positioning applications

JOURNAL=Frontiers in Physics

VOLUME=Volume 10 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/physics/articles/10.3389/fphy.2022.1058474

DOI=10.3389/fphy.2022.1058474

ISSN=2296-424X

ABSTRACT=The marine environment is extremely complex, which may cause scattering of underwater acoustic signals and seriously interfere with the reception. To extract more effective information from underwater acoustic signals, we use four types of multi-scale entropies, including multi-scale sample entropy (MSE), multi-scale fuzzy entropy (MFE), multi-scale permutation entropy (MPE), multi-scale dispersion entropy (MDE), to analyze and distinguish underwater acoustic signals. In this paper, two groups of the real-word underwater acoustic signal experiments were performed for feature extraction of ship-radiated noises (SRNs) and Marine ambient noises (MANs), the results indicated that the performance of MFE-based feature extraction method is superior to feature extraction methods based on other three entropies under the same number of features; and the highest average recognition rate (ARR) of MFE-based feature extraction method for SRNs reaches 100% when the number of features is 3.