摘要
针对非均匀杂波环境下自适应检测低慢小目标时,信号特征难提取,有效训练样本受限的问题,将低慢小目标建模为多维子空间模型,提出了基于多先验谱模型的低慢小目标子空间检测器构造方法。该检测器使用纹理分量为逆伽马分布的复合高斯模型来描述杂波,利用多先验谱模型的线性组合来表示杂波协方差矩阵的逆,能在均匀和非均匀杂波背景下检测低慢小目标。仿真表明,该检测器比传统基于渐进最大似然估计协方差矩阵的检测器以及单独基于多先验谱和子空间的检测器性能更好,并且训练样本数不足的情况下保持很好的性能。
When adaptively detecting the Low-atitude Slow and Small(LSS)targets in inhomogeneous clutter with limited number of training samples,the extract of radar signal feature is difficult.Aiming at this problem,a detector is proposed by combining multiple a-priori spectral models with multidimension taget subspace models.This paper models the heavy-tailed clutter as compound-Gaussian process with inverse gamma distributed texture and assumes that multiplea-priori spectral models can be obtained and the inverse of the clutter covariance matrix is the linearcombination of them.Then the Generalized Likelihood Ratio Test(GLRT)approach is used to propose three adaptive detectors.Simulations show that the proposed detector has better performance than the others,especially under the condition of limited number of training data.
作者
吕宽
张玉
唐波
LYU Kuan;ZHANG Yu;TANG Bo(School of Electronic Countermeasure,National University of Defense Technology,Hefei 230037,China)
出处
《火力与指挥控制》
CSCD
北大核心
2018年第9期182-185,共4页
Fire Control & Command Control
关键词
低慢小目标
目标子空间
复合高斯模型
多先验谱模型
LSS targets
taget subspace
compound-gaussian clutter
multiple a-priori spectral models
