Research Article
Underwater Information Sensing Method Based on Improved Dual-Coupled Duffing Oscillator Under Lévy Noise Description
@INPROCEEDINGS{10.1007/978-3-030-92635-9_10, author={Hanwen Zhang and Zhen Qin and Dajiang Chen}, title={Underwater Information Sensing Method Based on Improved Dual-Coupled Duffing Oscillator Under L\^{e}vy Noise Description}, proceedings={Collaborative Computing: Networking, Applications and Worksharing. 17th EAI International Conference, CollaborateCom 2021, Virtual Event, October 16-18, 2021, Proceedings, Part I}, proceedings_a={COLLABORATECOM}, year={2022}, month={1}, keywords={Underwater information sensing L\^{e}vy noise Dual-coupled Duffing oscillator}, doi={10.1007/978-3-030-92635-9_10} }- Hanwen Zhang
Zhen Qin
Dajiang Chen
Year: 2022
Underwater Information Sensing Method Based on Improved Dual-Coupled Duffing Oscillator Under Lévy Noise Description
COLLABORATECOM
Springer
DOI: 10.1007/978-3-030-92635-9_10
Abstract
Sensing underwater information has become particularly important to obtain information about the marine environment and target characteristics. At present, most interference models for underwater information sensing tasks under substantial interference choose Gaussian noise models. However, it often contains a strong impact and does not conform to the Gaussian distribution. Moreover, in the current research on the sensing of underwater unknown frequency signals, there are problems that the sensing method cannot sufficiently estimate the parameters of the unknown frequency signal, and the signal-to-noise ratio threshold is too high. An underwater environment sensing method is proposed by using the Lévy noise model to describe the underwater natural environment interference and estimate its parameters, which can better describe the impact characteristics of the underwater environment. Then, the intermittent chaos theory and variable step method are leveraged to improve the existing dual-coupled Duffing oscillator method. The simulation results show that the proposed method can sense weak signals in the background of strong Lévy noise and estimate its frequency, with an estimation error as low as 0.1%. Compared with the original one, the minimum signal-to-noise ratio threshold is reduced by 3.098 dB, and the computational overhead is significantly reduced.
