A Methodic Research On The Detection And Track Of Weak Maneuvering Targets

Modern aircrafts are usually highly mobile and able to perform stealthy activity,and their radar echoes are far weaker than traditional aircrafts.In order to realize the long-range detection and warning of these weak maneuvering targets in strong clutter background,it is necessary to improve the detection capability of radars to weak target signals.Apparently traditional techniques and methods cannot fully live up to these requirements.Therefore,the focus of this paper is to study the new detection and tracking technology,to propose new detection and tracking methods and to improve modern tactical weapons' detection and tracking ability on weak maneuvering targets in complex environments.First of all,this paper focuses on the detection and tracking of weak maneuvering targets,and studies the motion model,observation model and calculation model of typical weak maneuvering targets in depth.Secondly,this paper studies and applies the calculation method of particle filter and the realization method of bootstrap filter to the nonlinear problem,and effectively solves the problem of model mismatch caused by nonlinear non-Gaussian noise and fast random maneuver.Based on this idea,combined with the advantages of smoothing device,the filtering results removes the outliers(such as data glitches)using tree-based particle smoothing method.Then,four traditional weak moving target detection and tracking algorithms are studied,such as tracking algorithm based on Hough transform,tracking algorithm based on stochastic context-free grammar,tracking algorithm based on dynamic programming,tracking algorithm based on Bernoulli filter.In the mean time,simulation experiments are carried out.Finally,based on the analysis of the unbiasedness of particle filter smoothing method and four traditional weak maneuvering target tracking algorithms,this paper optimizes the strategy of filtering sampling,and proposes an improved Sequential Smoothing-Particle Filter,ISS-PF).It not only solves the problem of high cost,but also improves the accuracy of filtering,and can effectively track weak maneuvering targets.Both theory and practice have proved that with potential improvements compared with traditional methods,the newly proposed method has many advantages.The improved Sequential Smoothing-Particle Filter system can accurately capture the motion trace of the weak target,effectively reduce the observation error caused by the target maneuver,and update and process the data efficiently in real time.Through theoretical simulation,this paper also validates the existence of non-Gaussian noise and the mechanism of the interference of frequent maneuvering to the detection and tracking of the target and the traditional filter.