Research And Implementation Of Passive Location And Tracking Algorithm Under Non-common-view

With the continuous development of modern electronic technology,passive location and tracking system is developing towards the direction of high precision,low load,and fast speed,passive location and tracking technologies have matured and are used in actual reconnaissance operations.In the actual signal observation environment,there may be obstacles when multiple stations locate the target,which will cause the target signal not to be seen by all stations simultaneously,the scene is the non-consensus environment.In the multi-station nonconsensus scenes,due to the lack of information,the position of the target and the trajectory tracking of the target will have large errors.Secondly,the signal-to-noise ratio of the single observation environment decreases,and the amount of information decreases.These types of problem are collectively referred to as the missing information problem.Focusing on the missing information problem and non-consensus scenes,this paper studies traditional passive position methods and basic principles,direct position of array signals,direct position based on compressed sensing and passive target tracking filtering algorithms under non-consensus conditions.And this paper initially realizes the location and track platform based on the measured data.This paper proposes corresponding solutions to the non-consensus problems.First of all,this paper introduces the traditional passive position method of radiation source target,derives the algorithm principle and model,and simulates theoretical positioning performance.Aiming at the disadvantages of two-step data positioning algorithms,the positioning error changes with the accuracy of the measured variables,and the target informations are partial missing that cannot be used for accurate positioning in the actual observation environment,the paper studies the multiple signal classification(MUSIC)direct position algorithm based on array antennas,and the simulation shows that the positioning error is reduced compared with the two-step position method,especially at low signal-tonoise ratio(SNR),and different numbers of targets can be located according to the different number of array elements.Under the non-common-view of multi-station,this paper uses the compressed sensing algorithm and combines sparse conditions to solve the target position for the array signal positioning model in order to solve the problem that some observing stations cannot receive the target signal to determine target location.This paper derives the sparse representation model and specific dictionary matrix for compressed sensing direct position determination(CSDPD)algorithm,and uses the applicable reconstruction algorithm orthogonal matching pursuit(OMP)algorithm.In general,CSDPD can efficiently and accurately recover sparse targets.In a non-consensus environment,when the number of visible observation stations is reduced and the SNR is low,CSDPD algorithm has higher positioning accuracy than the twostep data position and MUSIC direct position methods.In addition,a passive target tracking algorithm in the non-common-view is studied in this paper.This paper analyzes the changes in the amount of observation information under this condition,and balances the confidence in the predicted value and the estimated value by adjusting the observed process noise,while adding the process noise matrix coefficient in order to make the process noise more in line with the actual tracking situation.Then stable and correct tracking results are getted.The simulation verifies the feasibility of the algorithm in a non-consensus environment and can improve the accuracy of target tracking estimation.Finally,based on measured actual data,this paper studies the conversion relationship between different observation coordinate systems and the realization of extended kalman filter(EKF)algorithm based on time difference of arrival(TDOA)under the non-consensus scenes.In the actual environment,we fuse and sort the pulse description word(PDW)data and separat out the PDW information of different targets received by each observing station,and select effective information for target position.The simulation experiment inputs the actual collected data,the output results are good,and complete the positioning and tracking operation effectively.