基于高斯-马尔可夫模型的海洋环境辨识方法

Ocean Environment Identification Based on Gauss-Markov Model

在解决海洋中的信号处理问题时,海洋环境的复杂性和多变性严重影响了信号处理系统的性能。为了在信号处理之前辨识出海洋环境参数,并将其融入信号处理框架中,则有希望提高信号处理器的性能。将海洋环境建模为高斯-马尔可夫模型,利用声速梯度数据和由Kraken模型得到的声压场数据,结合扩展卡尔曼滤波器算法,实现了对海洋环境的辨识。利用模基处理方法,简正波传播模型,可以辨识模函数和水平波数,并能估计出基阵所在位置的声压场。仿真数据和模型数据吻合较好,说明算法能够较好地估计海洋环境参数,为滤波、检测、定位、跟踪等应用提供了基础。

The complexity and inconstancy of the ocean environment lead to seriously degrade the performance of signal processing system. If ocean environment parameters can be identified before signal processing and then incorporated into signal processing schemes, it is contemplated to improve overall processor performance. In this paper o- cean environment was modeled as a Gauss - Markov process. According to the sound velocity profile and the pressure field derived from the Kraken model, extended Kalman filter algorithm was used to perform ocean environment identification. By this approach, the modal functions and the horizontal wavenumbers can be identified based on a normal mode propagation model, and the pressure field at the receiver array can be estimated. Simulation results and model data were found to to be consistent, which indicates the well performance of the algorithm, thus providing a basis for the enhancement, detection, localization and tracking.