Research On High Precision And High Resolution Passive Location Algorithm

Passive location method uses single or multiple reconnaissance stations to intercept signals and obtains the position of the target by analysis.With advantages of long-range detection and great electromagnetic concealment,it has been a hot topic for research at home and abroad.At present,the high-precision passive localization algorithm in single target and high-precision and high-resolution passive localization algorithm for multi-objective are two important research directions.With the increasing complexity of battlefield environment,there are often multiple time-frequency spatial mixing radiation source signals in the observation area,which are similar in carrier frequency,pulse width,pulse repetition period and other parameters.All these make the signal sorting almost impossible and the traditional positioning method ineffective.At this time,it is a challenging problem to achieve high-precision and high-resolution positioning.In order to solve the above problems,this study discusses the following four parts combining with the single-station and multi-station passive location system: single-station and single-target passive localization algorithm with high precision,single-station and multi-target passive localization algorithm with high precision and high resolution,multi-station and single-target passive localization algorithm with high precision,multi-station and multi-target passive localization algorithm with high precision and high resolution.For the problem that current single-station localization algorithm has low precision,this study proposed a single channel equivalent time difference passive localization algorithm based on the mathematical model of the pulse train emitter signal with fixed pulse frame period.This algorithm fully used the flexible characteristics of the small and micro reconnaissance platform,completed multiple sampling of the target signal at different positions of the trajectory,and then utilized the passive synthetic aperture array technology to effectively solve the problem of limited aperture of the reconnaissance platform.In the case of known and unknown pulse frame period,the algorithm estimated the equivalent time difference of the coherent pulse series radiation source signal,and derived the corresponding localization algorithm and Cramer-Rao bound in theory.The simulation results show that,compared with the uniform real linear array algorithm,the algorithm has higher positioning accuracy and can achieve high-precision positioning of the given target.To solve the problem of insufficient resolution for existing single-station and multi-target localization algorithms,this study proposed a single-station and multi-target direct position determination with double channels and a single-station and multi-target movement compensation-based direct localization algorithm based on Doppler frequency change rate respectively.The first algorithm took the emitter signals from coherent narrow-band radar as the research object,eliminated the phase effect of the time term by interferometric method along track,and then obtained the signal sequence containing the distance term between the reconnaissance station and the target emitter at different times.Then the radial distance history was compensated by the central point position correction of the observation area and Taylor series expansion,and the Doppler frequency and Doppler frequency change rate were estimated with high-accuracy by the fractional Fourier transform,and finally the location of target emitter was obtained.The second algorithm established the mathematical observation model of moving single station based on Doppler frequency and Doppler frequency change rate,and derived the Cramer Rao bound.The simulation results showed that both of the two algorithms could ensure the positioning accuracy and achieve high resolution,and solve the positioning problem of high-precision and high-resolution with multiple time-frequency spatial mixed radiation source signals in the observation area.For the problem that the estimated accuracy of time-frequency difference drops with the accumulation of time,an estimated algorithm using coherent pulse train based on relative motion compensation was proposed and then the analytic solution of emitter location was derived by using the estimated time-frequency difference.The algorithm used the center point of the observation area to construct a reference function to correct the radial distance difference.After that,through two-dimensional Taylor expansion,the higher order terms of the radial distance difference,such as acceleration and acceleration of acceleration,were transformed into functions of arrival time difference and Doppler frequency difference.Then generalized Keystone transform was used to eliminate the coupling between time difference and frequency difference.And the estimated values of time difference and frequency difference were obtained by two-dimensional Fourier transform.Finally,the analytic solution of the pseudo observation matrix was derived.The simulation results showed that compared to the frequency compensation algorithm,this algorithm could effectively overcome the problem that the estimated accuracy of time frequency difference deteriorates gradually with time accumulation,and the accuracy approached the Cramer-Rao bound when the number of collection pulses was high.To solve the problem that existing passive localization algorithm of multiple stations could not gain both high-precision and high-resolution simultaneously,two algorithms was proposed: the time domain direct position determination of fixed multiple stations for multiple targets and the time-frequency domain direct position determination of movement multiple stations for multiple targets.The first one took the deterministic unknown signal as the model,and established the corresponding movement compensation-based maximum likelihood direct position determination model for time domain respectively for the two cases of the known and unknown statistical properties of external noise,and deduced the Cramer-Rao bound of the localization algorithm when the signal and transmission loss were unknown.The second one firstly listed the deterministic unknown signal model received by multiple movement stations,and derived the direct movement compensation localization algorithm of the coherent pulse train under the background of white Gaussian noise,and the corresponding Cramer-Rao bound was given then.The simulation results showed that the two multi-station movement compensation-based direct position determination proposed in this study had better resolution while ensuring the positional accuracy,and could meet the locating requirements for multiple emitter signals aliasing in time-frequency and space-domain in the same observation area under low signal-to-noise ratio.