Time-evolving Sound Speed Profile Tracking Using Ambient Noises in Shallow Water

The time derivative of the noise correlation function has been proved to be approximately the sum of two time domain Green’s functions in opposite directions of propagation. This paper employs the ensemble Kalman filter algorithm to track time-varying sound speed profiles (SSPs) using ambient noises. As an auto-regressive process, coefficients of empirical orthogonal functions from historical SSPs are used as state vector and the frequency domain cross spectral density matrix is used as measurement vector. Tracking capabilities of this method were tested with simulated data generated by groups of SSPs measured during the SWellEx-96 Experiment. The results show that the proposed method enables continuous tracking of range-independent SSPs.