Research On Underwater Sound Source Localization Algorithm Based On TDOA

Underwater sound source localization technology is an important and indispensable technology for various underwater scenes,such as underwater robot detection,underwater navigation of unmanned boats,and ocean monitoring.In underwater sensor networks,we extensively use acoustic signals.This is because the attenuation of electromagnetic waves in water for terrestrial communications is very serious.Compared with electromagnetic waves,sound waves have good propagation characteristics,and they can cover several kilometers or even larger sea areas.The localization technology based on Time Difference of Arrival(TDOA)has good localization performance.This article focuses on the research and improvement of the underwater sound source localization algorithm based on TDOA.The main work is as follows:1.Aiming at the problem of poor time delay estimation and localization accuracy when TDOA localization is used alone.First,an improved orthogonal matching pursuit(OMP)algorithm is proposed,and a deep-sea channel model is constructed.The time delay difference of the sound wave propagation in a straight line is obtained through the time delay estimation of compressed sensing.Then,an improved three-step weighted least squares algorithm is used to locate the target.The simulation results show that the algorithm has good localization ability under noisy environment.On this basis,an underwater TDOA localization algorithm is proposed when the sensor position is floating.Aiming at the problem that the sensor will float around with ocean currents,resulting in the uncertainty of the sensor position.A twostage weighted least square algorithm is used to realize the localization of underwater targets.The simulation results show that the proposed algorithm still has better localization accuracy when the sensor position noise is large.2.Aiming at the problem of poor localization velocity estimation accuracy of Time Difference of Arrival and Frequency Difference of Arrival(TDOA/FDOA).First,a joint localization method of underwater TDOA/FDOA based on cross-ambiguity function is proposed,a cross-ambiguity function based on curved surface interpolation optimization theory is designed,and an optimization model with equality and inequality constraints is constructed.The penalty function method is used to turn the equality constraint problem into an unconstrained problem.On this basis,an underwater TDOA/FDOA joint localization method based on the Taylor-weighted least squares algorithm is proposed.The weighted least squares algorithm is first used to solve the rough estimated position and velocity of the target.Then the Taylor expansion of the TDOA/FDOA measurement value is solved.The localization error equation is constructed using the iterative method to continuously update the estimated target position and velocity.Finally,when the localization error is small enough or the maximum number of iterations is reached,the algorithm stops running and outputs the estimated target position and velocity.The simulation results show that these two algorithms can approach the Cramer-Rao lower bound under small noise conditions.3.Aiming at the problem that traditional localization algorithms based on time difference of arrival and angle of arrival(AOA)in underwater sensor networks have poor localization performance under high-angle noise,a two-stage weighted least squares algorithm using measured values and target position errors is proposed.In the first stage,the weighting coefficient matrix is obtained by introducing the measurement value error,which restrains the influence of the measurement value noise on the localization.In the second stage,a new localization equation is obtained by adding the target position error,and the target position calculated in the previous stage is adjusted by the solved target position error.The simulation results show that the proposed algorithm has good localization performance in the case of high-angle noise.