The Research In Radar Targets Tracking Under Complex Environment

Radar is an indispensable electronic device in modern warfare.Radar target tracking is a vital part of radar data processing,which is widely used in both military and civilian fields.However,the physical environment of the radar is very complex and there are a lot of clutter or false alarms in the received measurement data.In addition,sometimes there are situations where the number of targets is uncertain,which increases the difficulty of target tracking,especially the skywave over-the-horizon radar(OTHR).Therefore,there is an urgent need to study radar target tracking methos in complex environments.The research in this thesis was carried out under above-mentioned topic background,the main works are as follows:First,this thesis analyzes the radar target state estimation algorithm in details.The filtering techniques from linear Gaussian systems,nonlinear Gaussian systems to nonlinear non-Gaussian systems are studided,and the applicable scenarios,advantages and disadvantages of various filtering techniques are analyzed.Second,in order to further improve the estimation accuracy of the Cubature Kalman filter in the nonlinear Gaussian system,an improved algorithm(SSGQKF)based on the spherical simplex rule and the Gauss-Laguerre integration rule is proposed.Theoretical analysis and simulation results show that the improved algorithm can have a good compromise between accuracy and computation load in high nonlineality systems.Third,in order to solve the inherent particle degradation and particle depletion problems of standard particle filter in nonlinear non-Gaussian systems,a corresponding algorithm is proposed.First,the second-order central difference filtering algorithm which has good numerical stability is used as the importance density function,then the difference evolution is used to optimize the resampling process.Research results show that the improved algorithm can effectively overcome particle degration and depletion problems,and improve particle utilization and state estimation accuracy.Fourth,the target tracking problem under clutter environment based on data association algorithm is studied.The algorithms include the nearest neighbor association,the probabilistic data association,the joint probability data association and the interacting multiple model probabilistic data association algorithm.Considering the maneuver of the target,on the basis of the fixed delay technology and amplitude information,an improved interacting multiple probabilistic data association algorithm is proposed.At last,the improved method is simulated and analyzed with the traditional interacting multiple model probabilistic data association algorithm.Fifth,aiming at the problem of multi-target tracking in complex environments such as the number of the targets is uncertain and the clutter is dense,the multi-target tracking method based on random finite sets is studied,and the second-order centeral difference filtering algorithm and the differential evolution menthod are introduced into the sequential Monte Carlo probability hypothesis density(PHD).Simulation result show that the improved algorithm has a good effect on the multitarget tracking problem in a complex environment.Sixth,when an over-the-horizon radar tracks the target in ground coordinates,the target dynamic equation is linear,while the measurement equation is nonlinear.Aiming at this characteristic of the target tracking of the OTHR,the two-step filtering algorithm and the square root cubature Kalman filtering algorithm are combined and applied to the OTHR tracking system.The improved algorithm provides better accuracy without the need to analytically calculate Jacobians.Then,the problem of multipath target tracking based on probabilistic data association and random finite set are also studided,and the SSGQKF algorithm is integrated into the Gaussian mixture multipath PHD.Finally,the effectiveness of the algorithm is verified by simulations.