A new detection model of ship-radiated noise signal

With the continuous improvement of ship building technology in various countries, the ship-radiated noise signal (SRNS) in the ocean is becoming weaker and weaker, which makes the detection of SRNS more and more difficult. To improve the detection technology of SRNS, the thesis used the detection method based on the chaotic oscillator. To determine the chaotic threshold, an improved refined composite multiscale fluctuation-based reverse dispersion entropy (IRCMFRDE) is proposed. To detect the SRNS with a lower threshold of signalnoise ratio (SNR), after the chaotic threshold has been determined, proposed a detection method of SRNS with variable scale and variable step-size intermittent chaos based on differential double-coupled Van der PolDuffing (DCV-D) oscillator. For signals with unknown frequencies, the lowest SNR that can be detected by the proposed method is 5.6165 dB and 7.3595 dB lower than that of the existing variable step-size intermittent chaos detection methods based on DCV-D oscillator and differential DCV-D oscillator respectively. When cosine simulated signals under different background noises are used for detection, the accuracy rate of the proposed method is 99.81% and 98.95% respectively. When the measured signals (MS) are used for detection, the accuracy rate of the proposed method is 99.95%, 99.53%, and 99.56%. These results confirm the superior performance of the proposed detection method. To better demonstrate the superior performance of the threshold determination and the proposed detection method, we provide some experimental results in the supplementary material.