Study On Data Association Method In Multitarget Tracking Of Ballistic Missile Denfense Radar

Nowadays, many countries in the world are making great efforts to develop techniques of ballistic missile and missile defense system. There is more and more countries provided with advanced ballistic missile defense system. In the meanwhile, the penetration technology is developing too: the measurements such as multiple warheads, multi-false-target jamming, high altitude decoy are taken widely. So, multitarget tracking technology of ballistic missile defense radar is confronted with great challenge. The data association problem of multitarget tracking is a very meaningful subject in dense clutter or multitarget environment.Under the related item background of missile attack defend confront simulation and efficiency evaluation, this thesis is the study on data association method of multitarget tracking, consists of two main sections:The first section is the study on nonlinear filtering method of missile traget tracking, sets up the filtering coordinate for missile target tracking; sets up the trajectory generate model and generates simulation trajectory; uses the motion model based on J₂ correction and measurement model of hybrid coordinates; for the nonlinear measurement model, uses the Unsented Kalman Filter instead of the conventional Extended Kalman Filter, gets better tracking effect.The second section is the study on data association method of multitarget tracking, consists of two points:Firstly, for single taget tracking in clutter, the thesis researches the algorithm theory and realization of Nearest Neighborhood Standard Filter (NNSF) and Probabilistic Data Association Filter (PDAF), designs simulation experiments, analyses and contrasts the performance of the two filters, uses PDAF for single target tracking in clutter, gets good tacking effect.Secondly, for multitarget tracking, the thesis researches the algorithm theory and realization of Joint Probabilistic Data Association(JPDA), designs simulation experiments of tracking parallel targets with little distance, cross targets with little angle, and missile target with little distance, all gets good tracking effects.