复杂海洋环境噪声下甚低频声信号检测方法

The detection of very low frequency (VLF) signal in the complex marine environment is studied to detect the submarine targets in the complex marine environment. The Lévy distribution is used to describe the noise in complex marine low-frequency environment, and the logarithmic moment method is used to estimate the distribution parameters of the measured noise. A priori knowledge is provided for the following system parameter selection. And then a stochastic resonance method is proposed to detect the very low-frequency signal in the complex marine environmental noise, and the second-order stochastic resonance method under the Lévy noise is analyzed. The matching parameters of stochastic resonance system under Lévy noise are derived, and a detection system of second-order bistable matching stochastic resonance is established. The simulated results show that the detection performance of the second-order matched stochastic resonance detector is much better than that of the traditional correlation detector. When the false alarm probability is 0.01, the detection probability can reach 0.9. Compared with the correlation detector, the required input signal-to-noise ratio can be reduced by 10 dB. Through the field lake experiment, it was verified that the system can effectively detect the VLF sound signals generated by underwater mobile vehicle.