Inversion Of Sound Speed Profile In Dynamic Ocean Environment

The ocean environment is dynamic and variable. where the propagation of acoustic signal is affected by various environmental factors. Among all the environmental parameters, the sound speed profile (SSP) in water column is highly variable in time and space. Moreover, it is sensitive to the sound propagation, thus affecting the performance of the systems for underwater communications and localization.Generally, the SSP can be directly measured by the instruments. The in-site measurements are not only time-consuming but also unable to obtain observed data in a wide range. By using the sound propagation model and the observed sound pressure data, the acoustic inversion technique provides a method which can invert the interested environment parameters rapidly and efficiently.According to the evolution characteristics of SSPs, this paper investigates the tracking techniques of time-evolving SSPs based on sequential filtering. The application of some algorithms on the SSP inversion is introduced, such as the extended Kalman filter (EKF), the ensemble Kalman filter (EnKF) and the particle filter (PF). In terms of the characteristics of nonlinear systems and non-Gaussian distributions in the SSP inversion problem, we develop the algorithm of the ensemble Kalman-particle filter (EnKPF) for the tracking of time-evolving SSPs. By numerical simulations and experimental data processing, the performances of these sequential filtering algorithms are analyzed and compared.Additionally, this paper puts forward a method to invert SSPs using a moving source, which tries to resolve the contradiction between the observation range and the resolution in a fixed network. By employing moving sources, we can increase the local sampling points and enrich acoustic/environment observation information to improve the accuracy and resolution of the estimated parameters. Because the source position information is also included in the observed data, the parameters of the source position can also be estimated using this method at the same time. Finally, the feasibility of this method is verified through numerical simulations and experimental data.