Research On Localization Algorithm Based On TDOA/FDOA/DDR Measurements For Moving Target

In passive localization,the observation stations are used to obtain the measurements such as Angle of Arrival(AOA),Time of Arrival(TOA)and Time Difference of Arrival(TDOA)of the target signal,so as to determine the target's position.The researchers introduced the measurement of frequency difference of arrival(FDOA),that the localization algorithm based on TDOA and FDOA can estimate the position and velocity of the target.In addition,measuring Differential Doppler Rate(DDR)can compensate for Relative Doppler Companding to improve the estimation accuracy of FDOA,and can be used for localization of target.Currently,TDOA,FDOA,and DDR-based localization algorithms consist of two steps:the first step is to estimate the target roughly.the second step is to further optimize the target localization results based on the rough estimate.However,the rough estimate of this kind of algorithm is calculated under the precondition that the variables are mutually independent,which neglects the pertinency between variables,leading to insufficient rough estimation accuracy under high noise,and the final estimation accuracy of the target in the second step is reduced.To solve the above problem,an improved localization algorithm using Lagrange multiplier is proposed in this paper.A set of TDOA,FDOA and DDR equations were obtained by analyzing the relationship between the observation stations and the target,and then the equations were linearized by introducing redundant variables.The localization problem is constructed as a constrained weighted least squares problem and converted to an unconstrained minimization problem by using Lagrange multiplier.Finally,this problem is solved by eigenvalue decomposition and polynomial rooting operation to get the rough estimation of target.The improved rough estimation method has better prediction accuracy because it abandons the unreasonable assumption that the variables are independent of each other.Finally,the constraints are linearized and the rough estimation is substituted into the Weighted Least Squares solution to obtain the final localization result of the target.Simulation results show that the improved estimation algorithm proposed in this paper has better localization accuracy.