非均匀强杂波下的目标检测问题研究

发布时间：2018-07-03 05:49

  本文选题：非均匀强杂波 + 微弱目标检测　； 参考：《电子科技大学》2014年博士论文

【摘要】：杂波背景中目标检测是雷达系统的基本任务之一。随着现代隐身技术的发展,目标的雷达截面积显著减小,回波信号变得十分微弱,信杂比显著下降;同时,城市、山地、和海浪等背景产生的杂波,强度大,均匀性差。这两方面因素的共同作用的结果,导致基于均匀杂波假设的传统检测器性能明显下降,尤其是在低信杂比或强杂波情况下,性能下降更为显著,这使得非均匀强杂波下的目标检测面临严峻挑战,已成为现代雷达信号与数据处理必须解决的难点问题。本文围绕非均匀强杂波下的目标检测问题,开展了杂波建模、检测器设计、仿真检验和实测数据验证等研究工作,主要内容如下:1.针对内海、湖泊等逆高斯(IG)调制复合高斯(CG)(IG-CG)非均匀强杂波,提出了基于两步广义似然比(GLRT)准则的自适应检测器,克服现有非均匀检测器杂波模型失配的缺点,检测性能得到改善。2.针对纹理分量部分相关的非均匀地/海强杂波,根据两步广义似然比(GLRT)准则,提出根据纹理分量和斑点分量估计杂波协方差矩阵,以改善检测器的杂波自适应性能,能够降低杂波模型失配对检测性能的影响。3.针对城市、海洋、植被等复合高斯非均匀强杂波,提出了GLRT-MSD,Rao-MSD和Wald-MSD等三种自适应多帧检测器,以期有效利用目标杂波的帧间相关性差异,提升分辨单元内运动小目标的检测性能。4.针对循环平稳的非均匀海杂波,提出了M-NHD和SV-NHD非均匀多帧检测器,可以避免非均匀参考数据对分辨单元内运动目标检测性能的不利影响。5.针对城市、海洋、植被等复合高斯非均匀强杂波,提出VL-HSCD,VL-HKelly和VL-HAMF等自适应非均匀多帧检测器,结合混合协方差矩阵估计方法和Viterbi-like(VL)帧间积累方法,可以提升跨分辨单元运动目标的检测性能。上述的检测算法,已通过仿真数据或实测数据的验证,其中,实测杂波数据为国际通用的IPIX雷达数据,仿真中的杂波参数主要来自于对实测数据的估计。
[Abstract]:Target detection in clutter background is one of the basic tasks of radar system. With the development of modern stealth technology, the radar cross section of the target decreases significantly, the echo signal becomes very weak, and the signal-to-clutter ratio decreases significantly. At the same time, the clutter produced by the background of city, mountain, wave and so on is of great intensity and poor uniformity. As a result of the combined action of these two factors, the performance of traditional detectors based on the assumption of uniform clutter is significantly reduced, especially in the case of low signal-to-clutter ratio or strong clutter. This makes the target detection under non-uniform strong clutter face a severe challenge and has become a difficult problem that must be solved in modern radar signal and data processing. Focusing on the problem of target detection under non-uniform strong clutter, this paper has carried out research work on clutter modeling, detector design, simulation and verification of measured data. The main contents are as follows: 1. In this paper, an adaptive detector based on two-step generalized likelihood ratio (GLRT) criterion is proposed for inhomogeneous strong clutter modulated by inverse Gao Si (IG) modulation in lakes and lakes, which overcomes the shortcomings of existing heterogeneous detector clutter models. Detection performance improved. 2. 2. Based on the two-step generalized likelihood ratio (GLRT) criterion, the clutter covariance matrix is estimated based on texture component and speckle component to improve the adaptive performance of the detector. It can reduce the influence of clutter model mismatch detection performance. In this paper, three adaptive multi-frame detectors, GLRT-MSD Rao-MSD and Wald-MSD, are proposed to improve the detection performance of moving small targets in the resolution unit by effectively utilizing the inter-frame correlation difference of target clutter and improving the detection performance of moving small targets in the resolution unit. For cyclic stationary heterogeneous sea clutter, M-NHD and SV-NHD non-uniform multi-frame detectors are proposed, which can avoid the adverse effect of non-uniform reference data on the detection performance of moving targets in the resolution unit. For urban, oceanic, vegetation and other complex Gao Si heterogeneous strong clutter, an adaptive nonuniform multi-frame detector, such as VL-HSCDNF-VL-HKelly and VL-HAMF, is proposed, which combines the mixed covariance matrix estimation method and the Viterbi-like (VL) inter-frame accumulation method. It can improve the detection performance of moving targets. The above detection algorithms have been verified by simulation data or measured data. Among them, the measured clutter data are international IPIX radar data, and the clutter parameters in the simulation mainly come from the estimation of the measured data.
【学位授予单位】：电子科技大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TN957.52
，

本文编号：2092632