Narrowband Radar Imaging For Midcourse Targets Based On Nonideal Scattering Characteristics

The narrowband radar is mainly used for the early warning of long-range targets.Compared with wideband radar system,narrowband radar system has many advantages in target tracking,detection range,micro-Doppler feature extraction and so on.The researches on high-resolution narrowband radar imaging have been developed since the 1980 s,and the tomographic techniques and imaging methods based on time-frequency domain are proposed for narrowband imaging with wide obversation angle.The advantages of narrowband radar system in target detection range contribute to early detection and early recognition of the midcourse target.Thus,narrowband radar imaging for midcourse targets is a new technical approach in aerial targets imaging and recognition problem.As the development of the research in nonideal scatterring characteristics of midcourse targets,the micro-Doppler of scattering centers does not satisfy with sinusoid hypothesis,which embarrasses the previous narrowband imaging methods.This thesis aims to further improve the ability of narrowband radar in imaging,character inversion and targets recognition for midcourse targets,and enlarge its superiority in targets detection.According to the Doppler analytical model for nonideal scattering centers of representative midcourse targets with precession motion,this thesis mainly focuses on the breakthrough of two kinds of narrowband imaging methods based on time-frequency characteristics and tomographic techniques,using mathematical tools such as n-dimensional general Radon transform and its inverse transform.Besides,the translational compensation method for narrowband imaging is also studied,which improves engineering practicality of the research results.The breakthrough of this research project will bring new technical means for the recognition of midcourse targets.Chapter 1 illustrates the research background and the significance of this research project,and introduce the research status of narrowband radar imaging from two aspects: the tomographic technique and imaging methods based on time-frequency domain;the research status of target nonideal scattering characteristics and translational motion compensation is also introduced.Then the present investigation problems of narrowand radar imaging for midcourse target are summarized,after which the research content and the organization chart of this thesis are presented.Chapter 2 mainly illustrates the nonideal scattering phenomenon of midcourse targets and establishes the micro-Doppler model of nonideal scattering centers.Then the micro-Doppler characteristics of the midcourse target are analyzed by the simulations and experiments,which lays the foundation for the further research of midcourse targets imaging based on nonideal scattering characteristics.Chapter 3 proposes a narrowband imaging algorithm via complex-valued empiricalmode decomposition-time frequency distribution-general Radon transform(CEMD-TFD-GRT),which eliminates the suppress interference of specular reflection on midcourse target imaging.Firstly,the principle of CEMD-TFD-GRT algorithm is introduced and the imaging coordinates are deduced for both ideal-type scattering centers and sliding-type scattering centers.Then,the computation complexity of CEMD-TFD-GRT is analyzed and the Cramer-Rao low bounds for the coordinates estimation are derived.Finally,the experiment results with anechoic chamber data demonstrate that the scattering centers of precessional targets can be imaged with the proposed algorithm even when specular scattering occurs.The noise influence on the proposed algorithm is also presented with the experiments.Chapter 4 proposes a method to estimate the dimensions of midcourse targets based on time-frequency characteristics.The CEMD algorithm is utilized to separate the micro-Doppler curves and then the scattering centers on the midcourse target are classified based on the deviation degree of micro-Doppler from sinusoid.The threshold for the classification is determined with the simulation data.After the classification,the target dimensions are estimated according to the relationship between the micro-Doppler parameters and the target dimensions.The simulations and experiments validate the effectiveness of the proposed method.The influence of motion parameters estimation errors and the noise on the estimation of target dimensions is also illustrated with simulation data.Chapter 5 studies the complex-valued back projection(CBP)imaging algorithm based on nonideal scattering characteristics of midcourse target.The principle of conventional CBP algorithm based on sinusoidal phase is introduced first and the influence of nonideal scattering of midcourse target on CBP imaging is analyzed.Then,an improved CBP algorithm for three dimensional imaging is proposed based on the sliding-type scattering centers characteristics.This proposed algorithm can image the target in the local coordinate system,which does not require the image calibration.The simulation results demonstrate the effectiveness of the proposed method.Chapter 6 studies the translational motion compensation techniques of narrowband radar imaging,and proposes a nonparametric method to compensate the residual translational motion(RTM)based on the Doppler spectrums.The influence of RTM on narrowband radar imaging is analyzed first.Then,the procedure of the proposed compensation method is presented.This method combines the Viterbi algorithm and empirical mode decomposition(EMD)algorithm to estimate the instantaneous frequency of dominant scatterer and extract the Doppler frequency trend induced by RTM.The residual translational phase can be retrieved by integrating the extracted Doppler frequency trend.The simulation results validate the effectiveness of the proposed method.Chapter 7 summarizes the research work and concludes the innovations of this thesis,then indicates the deficiencies and point out the future work to be studied.