Tracking of the Time Evolving Sound Speed Profiles Using Ensemble Kalman Filtering

Abstract

Environment uncertainties lead to difficulties for accurate underwater acoustic field predictions. This paper presents a synthetic test to track the time evolving sound speed profiles (SSPs) using acoustic pressure field, which is derived by Princeton Ocean Model (POM) outputs based on the representative environmental model of a sea trial. In the simulation, range independent SSPs are parameterized by Empirical Orthogonal Functions (EOFs) and their coefficients, which are placed into the state-space form. Their variations are modeled as a three-order Auto-Regress (AR) process, and tracked by an Ensemble Kalman Filter (EnKF) from observing the underlying sound fields as generated by a parabolic equation model. Promising tracking results are obtained from the synthetic data, especially for the first two orders coefficients, suggesting the feasibility of EnKF to assimilate the states with acoustic measurements. The performance of the tracking algorithm with respect to Signal to Noise Ratio (SNR) of measurements, the samples and the amount of observations as applied in EnKF are also discussed in this paper.