Reverberation is an important ocean phenomenon that involves bottom properties. This paper proposes a method of geoacoustic inversion using bottom reverberation in the deep ocean. Models for the probability density function, including Rayleigh, Lognormal normal, K, and Weibull distributions, are commonly used to describe the envelope of bottom reverberation. Based on the measured data in South China Sea, the envelope of bottom reverberation in the experiment is consistent with K distribution. A double exponential distribution is used to generate the amplitude of reverberation, of which the envelope obeys the K distribution. In addition, the propagation and scattering models are introduced as the parameters of double exponential function, wherein the model has a physical mechanism. Therefore, a bottom reverberation model with both physical and statistical characteristics is suitable for geoacoustic inversion. The cost function is used to minimize the mean square difference between the measured and modeled envelope of the bottom reverberation. The inversion results are consistent with those from previous research.

中文翻译：

---

深海中基于物理-统计底部混响模型的地声反演

混响是一种重要的海洋现象，涉及海底特性。提出了一种利用深海底部混响进行地声反演的方法。概率密度函数的模型（包括瑞利分布，对数正态正态分布，K和威布尔分布）通常用于描述底部混响的包络。根据南海的实测数据，实验中底部混响的包络与K分布一致。使用双指数分布来产生混响幅度，其中包络服从K分配。另外，引入了传播和散射模型作为双指数函数的参数，其中该模型具有物理机制。因此，具有物理和统计特性的底部混响模型适用于地声反演。成本函数用于最小化底部混响的测量和建模包络之间的均方差。反演结果与以前的研究结果一致。