Research On Passive Localization Based On Doppler Frequency Shift Of Multiple Moving Receivers

In the past few decades,passive localization has played an important role in navigation,sonar,radar,electronic countermeasures,wireless communication and other fields.Passive localization includes the localization of source and the localization of the external transmitter,that is,the system itself does not transmit signals,but uses the signal emitted by the source or the signal reflected by the target for localization.In electronic reconnaissance,when the signal bandwidth is insufficient to cause the blurred time difference,or the phase calibration and time synchronization are limited by the complexity of the system.The target position estimation is more inclined to utilize Doppler frequency shift information.This technology can make up for the above defects and has lower requirements on the receiving system equipment.Coupled with the rapid development of frequency technology and the increasing accuracy of frequency measurement,passive localization based on Doppler frequency shift has great development potential.However,since the Doppler frequency shift is more complex to form the nonlinear equations with the target state than other positioning parameters,it is difficult to estimate the target position by using Doppler frequency shift.This paper mainly studies the multi-station passive location problem based on Doppler frequency shift in diverse scenarios.The corresponding localization algorithms are proposed for different scenarios and the performance of the proposed algorithms are verified by theoretical analysis or numerical simulation.The main research contents and contributions of this dissertation are as follows:1.The passive source localization problem based on Doppler frequency shift by multi-receivers is studied.Firstly,a received signal model including Doppler frequency shift is established.Two positioning algorithms are proposed under this signal model,which solves the problem of the huge computational complexity of the grid search method and the requirement of iterative algorithm on an initial value with a certain accuracy.One method is the explicit solution algorithm.By introducing intermediate variables,the final estimate of the source position is written as an explicit expression.The calculation of this method is simple without divergence problem and the computational power requirement for the system is low.The explicit estimation result can be used as the final positioning result,or the initial value of the iterative algorithm to further improve the localization performance.Another is the Semi-definite Relaxation(SDR)algorithm,which transforms the original non-convex estimation problem into a convex optimization problem by using the relaxation processing technique.This algorithm has a better estimation performance than the explicit solution algorithm and is close to the Cram?er-Rao lower bound(CRLB).2.The passive source localization problem based on Doppler frequency shift by multi-receivers in the presence of receiver position errors is studied.The CRLB with receiver position errors is deduced and compared with the one without receiver position errors.The influence of Doppler shift measurement errors and receiver position errors on the positioning result is analyzed.Based on the requirements of fast positioning and precise positioning,corresponding positioning algorithms with the ability to handle the receiver position errors are proposed.The simulation results proved that the proposed algorithm has higher estimation accuracy than the one without considering the receiver position errors in the presence of receiver position errors,and has the ability to deal with the receiver position errors.3.The passive source localization problem based on Doppler frequency shift with earth radius constraint by multi-receivers is studied.For locating the source that is on the surface of the earth with elevation information,this dissertation analyzes and derives the CRLB under the spherical constraint.This CRLB is utilized as a benchmark of the localization performance.Then,the constraint is introduced into the solution method so that the indirect semi-definite relaxation(WLS-SDR algorithm)and direct semi-definite relaxation(SDR algorithm)method are proposed to estimate the source position.The performance of the two algorithms is compared.The simulation results show that the direct semi-definite relaxation solution has higher positioning accuracy.The location model with receiver position errors is deduced and optimized so that a positioning method to correct this kind of errors is proposed.4.The problem of direct position determination of Doppler frequency shift based on external transmitters is studied.The model of the received signals is established in the condition that the system contains a reference channel.Then,the CRLB of the target localization is deduced and a direct positioning algorithm with reference signal is proposed.Finally,the direct position determination algorithm is compared with the two-step localization method under the condition of different noise levels,different number of observations,different number of receivers and different number of snapshots.The simulation results show that the performance of the proposed direct position determination algorithm is better than that of the two-step localization method under weak signals conditions.5.The problem of direct position determination of Doppler frequency shift based on external transmitters is studied in the case where there is no reference channel and both the receivers and transmitters are in motion.Firstly,the received signal model without reference signal is established,and the corresponding CRLB is derived.Then,the direct position determination algorithm without reference signal is proposed to transform the maximum likelihood estimation of the target position into the cost function in the form of eigenvalues.Simulation results show that the accuracy of the proposed direct position determination algorithm is higher than that of the two-step localization method,especially in low signal-to-noise ratio and fewer sample points conditions.