Study On Data Assimilation For The Acoustic Field In The Ocean

Sound speed profile is an important physical parameter in the study of underwater sound propagation,communication and signal processing.There are many ways to get the sound speed profile information such as through the ocean dynamics model,direct measurement or the inversion of ocean acoustic propagation model.Data assimilation can make the different kinds of information into integration and improve the accuracy of the sound speed profile estimated.Sound speed profile estimating based on the travel time of sound propagation is widely used because of the advantage that it is faster for processing and less affected by the seabed acoustic parameters.However,in practical applications,the measurement time of sound propagation often has some error from technical or human aspects.This kind of uncertainty would make adverse effects on the inversion results of the sound speed profile.Data assimilation methods of ocean observing were studied in this paper.Sea surface temperature parabolic regression prediction model had been studied and an improvement was made based on it with data assimilation methods.Based on the uncertainty of acoustic measurement and acoustic propagation model,the sound speed profile data assimilation processing system with three-dimensional variational algorithm was established in this paper.An analysis about the effects on the accuracy of sound speed profile inversion impact from the measurement errors had been made through Matlab simulation.And a correction for the inversion results with measurement errors and uncertainty had been made by data assimilation system.Sound speed profile correction results show that data assimilation method could give a more accurate estimation of sound-speed field.Data assimilation methods provide a new idea for acoustic data processing and inversion of sound speed profile.