Title: Non-Gaussian ocean ambient noise model with certain kurtosis

Author(s): SONG Guoli; GUO Xinyi; MA Li; LI He;

Affiliation(s): Key Laboratory of Underwater Acoustic Environment, Chinese Academy of Sciences; Institute of Acoustics, Chinese Academy of Sciences; University of Chinese Academy of Sciences

Abstract: Based on sampling damped sinusoids uniformly, a non-Gaussian ocean ambient noise sequence with specific spectral characteristics and a specific kurtosis value is generated for the non-Gaussian ocean ambient noise simulation problem. Under the theoretical simulation conditions, it is found that when the kurtosis value is higher than 3.0 better simulation results can be obtained. But when the kurtosis value is less than 3.0,a method of down sampling or reducing the filter order should be adopted for the simulation implementation. A simulation analysis of ambient noise data from four ocean acoustic environments is given, including quiet environment noise data, ship noise data, air gun data and impulse like data, and the results show that under the above scenarios, the non-Gaussian ocean ambient noise with the goal spectrum characteristics as well as statistical characteristics can be produced, and the kurtosis value simulation error is less than 2.4%.