Research On Single-observer Multiple-Emitter Passive Coherent Location And Tracking Based On Angle And Time Delay Estimation

Single observer multiple radiation sources localization systems has been paid much more attention in the modern battlefield circumstance,because the system could catch information on a number of external radiation to detect target,and it has the advantage of self-hiding,far-distance detection,simple structure and easy to engineering.In this paper,under the background of the single observer and multiple radiation sources positioning system,the direction of arrival(DOA)estimation method,the time-delay estimation(TDE)method and the joint DOA and time difference of arrival(TDOA)positioning and tracking method are studied.The main works and achievements overview paper are as follows:1.A novel passive TDE method based on Markov chain Monte Carlo(MCMC)algorithm is proposed in this paper.The Maximum Likelihood(ML)estimation model of passive time delay is primarily built.Then we derive a fast algorithm of ML solution.An accurate time delay estimator using MCMC algorithm is proposed because the fast algorithm could only get the time delay which is the integer multiple of sampling interval.The Cramer-Rao lower bound(CRLB)of this model is also derived.Finally,the simulation analysis shows that the algorithm of this paper could obtain an accurate estimate of the time delay without the need for interpolation and initial value.The computation complex of the ML solution is lower than traditional iterative algorithm.2.A new time delay estimation based on MCMC algorithm in multipath environment is proposed.The ML model of passive multipath time delay estimation is built.Based on the characteristic of the likelihood function,restraint condition is increased to make the maximum likelihood(ML)functions have only one maximum value.MCMC algorithm is utilized to find the ML solution,after which we derive the CRLB of multipath time delay estimation.The simulation analysis shows that the mean square error(MSE)of the proposed algorithm can close to CRLB in the case without interpolation.3.The TSE TDOA positioning algorithm of single-observer multiple transmitters system is derived.Firstly,assume that the target location coordinates and distance from the target to the stations is not relevant,the ML estimation algorithm is iterated to get target position and distance estimation.Then the ML algorithm is also used to remove the assumption and get accurate target location.We also derive the CRLB of this algorithm and the covariance matrix of estimation errors.The simulation shows that,when the TDOA error is small,the estimated MSE closes to CRLB.4.Joint DOA and TDOA ML positioning method is proposed,because the MSE of estimation location of TDOA positioning method deviate CRLB.Joint DOA and TDOA to construct the ML function of the target location.And then Newton iterative method is used for solving the ML solution to the target position estimation.The initial iteration value is given by the least squares(LS)algorithm.The CRLB of this algorithm and the covariance matrix of estimation errors are also derived.The simulations prove that the CRLB of positioning system joint DOA and TDOA is lower than that of TDOA system,the estimated MSE of proposed algorithm closes to the CRLB.5.Because LS algorithm could not obtain precise target location,the constrained total least squares(CTLS)algorithm is used to get accurate target position.This algorithm with the objective of minimizing the overall error,then build the CTLS algorithm objective function,and then using Newton iterative algorithm and the general iterative algorithm to achieve the target position estimation,after which the estimation error covariance matrix is also derived.The simulation analysis shows that these two iterative algorithms in this paper are equivalent,and the MSE of proposed method closes to CRLB.6.According to the pseudo-linear positioning method,we propose the pseudo-linear Kalman filter(PKF)algorithm for moving target tracking.According to the relationship of the measured value and the actual value,the Taylor expansion method is utilized to establish the actual observation equation.And then the traditional Kalman filtering method is used to achieve target tracking.Because PKF algorithm model contains the unknown target location,the iterative PKF(IPKF)algorithm is proposed to track moving targets.The simulation analysis shows that the accuracy,the stability and convergence rate of PKF tracking algorithm of this paper is higher than the EKF algorithm,and IPKF algorithm tracks target more accuracy and astringes more quickly than PKF algorithm.