Research On Inversion Of Sound Speed Profile With Sources Of Opportunity

Acoustic waves are the main carrier used in underwater communication.The propagation of acoustic signals in the ocean is influenced by various marine environmental parameters,and it is particularly sensitive to the distribution of sound speed profile(SSP)in the water body.Ocean acoustic tomography provides an effective method for large-range and multi-scale continuous ocean observation by combining acoustic propagation model with actual observation data.Ambient noise is the background interference field for the processing and application of acoustic signals.However,it is also the inherent sound field in the ocean,which contains the characteristic information of the marine environment.Extracting and reconstructing feature information using ambient noise provides a novel idea for acoustic tomography.Based on the shipping noise,this paper studies the method for the space-time tracking of SSP.The time domain Green’s function between two points during sound wave propagation can be recovered from the noise cross-correlation function,and SSP can be inverted according to its structure.Due to the complexity and variability of the marine environment,SSP has typical characteristics of spatio-temporal variation.This paper tracks SSP based on the idea of sequential filtering,starting from the statistical characteristics obtained from the prior information to construct a state evolution pattern followed by the sound field.In the time-evolving SSP tracking of the vertical section,the ensemble Kalman filter algorithm is used to realize the sequential estimation of the sound field.For the state-space model in the algorithm framework,this paper models the state equation describing the evolution characteristics of the sound field as an autoregressive process,and uses the forward model for obtaining the frequency domain cross spectral density information of ambient noise as the measurement equation.This paper studies the space-time tracking methods of the sound field on basis of the vertical section SSP inversion.The development and application of underwater acoustic sensor network(UASN)has expanded the range of ocean observation and enriched the acoustic environmental observation information.The UASN consists of multiple sensor nodes placed inside the observation area,therefore,the entire area is divided into several triangles according to the node topology.The sub-triangles are divided again in the triangles,and the sub-triangles are used as the space tracking unit to model the variation of SSP as a space-time autoregressive process.This paper combines the information filter algorithm suitable for linear systems with the ensemble Kalman filter algorithm through statistical linear error propagation approximation,and applies it to the UASN,studying the centralized and distributed spatio-temporal ensemble Kalman information filter algorithm to realize the space-time tracking estimation of the sound field.This paper exploits the data provided by SWellEx-96 experiment to verify the feasibility of the ensemble Kalman filter algorithm and ensemble Kalman information filter algorithm to track the time-evolving SSP.Using the data of the South China Sea simulated by the regional ocean model system,the performance analysis of the centralized and distributed spatio-temporal ensemble Kalman information filter method in 3D SSP tracking is conducted under different scenarios.The method proposed in this paper can be applied to space-time tracking of the sound field under the framework of acoustic sensor network based on sources of opportunity.