Study On Warhead Motion Model And Its True-false Discrimination In MIMO Radar With Distributed Antennas

With the rapid development of the worldwide military weapon technology and electronic warfare technology, in the increasing complicated international political situation, the study of distinguishing ballistic missile warhead target has become increasingly active in many countries. In this thesis, MIMO(Multi-Input Multi-Output) radar is employed to investigate warhead target parameter estimation and motion feature analysis, with the subject of solving warhead target recognition problem. This thesis mainly involves MIMO radar signal model with widely separated antennas, warhead target motion parameter estimation performance and description in the state-space model, and finally a new local optimum detection-- LOUD(Locally Optimum Unknown Direction) algorithm is applied to detect target state characteristics. After the contrast with LRT(Likelihood Ratio Test) algorithm, simulation shows that the LOUD method has fast detection and high sensitivity to small changes in target characteristics. The main content of this thesis is divided into four parts:Part I: MIMO signal modeling and parameter estimation with distributed antennas;It introuduces the principles, characteristics and patterns of MIMO radar system, and mainly studies the basic principles of MIMO radar with distributed antennas as well as the signal model at the transmitters and receivers. On basis of the MIMO radar receving signal model, target parameter estimation method is derived.Part II: build warhead target micro-motion modelIn order to analyze the differences and characteristics of the true false warhead, the motion model of the warhead target flying in the mid-course phase is decomposed to translation model and the micro-motion model. Then micro-motion effect of warhead in mid-course phase is modelled in mathematical description and analyzed.Part III: formulate the state space model for the warhead target motion model;According to the linear system theory and analysis of the ballistic target motion characteristics, the target motion state transfer equation of target and the observation equation with MIMO radar are established, then ballistic target motion state space model is built.Part IV: application of hypothesis testing theory in the warhead target recognition;A brief outhline of the general hypothesis testing theory is presented, and the warheads binary complex hypothesis testing problem is built by reflecting the difference between true and false warheads on the subtle change in the coefficient matrix of the state transfer equation. Then the LOUD algorithm is employed for local change detection in the binary hypothesis test of warhead target recognition. Compared to the likelihood ratio test in the ideal case(Ideal Likelihood Ratio) and the generalized likelihood ratio test(GLR) algorithms, LOUD test is simulated and analyzed in the case of detecting local change.