一种新的高分辨雷达杂波模型参数估计方法  

A Novel Parameter Estimation Method of Clutter Model for High Resolution Radar

作　　者：姜斌[1] 任双桥[2] 黎湘[2] 郭桂蓉[2] 

机构地区：[1]通信指挥学院,武汉430010 [2]国防科技大学四院空间信息技术研究所,长沙410073

出　　处：《现代雷达》2007年第5期14-18,共5页Modern Radar

基　　金：国防预研项目(No.4130300203)资助;国家自然科学基金(No.60402032)资助

摘　　要：高分辨率雷达体制下,杂波散斑分量与调制分量更适合用广义Gamma分布模型描述。该文首先建立了一种适用范围广泛的广义复合概率分布模型,并推导了其统计特性。然后研究了最小二乘参数估计方法,并以参数解耦和充分利用杂波序列信息为突破点,提出了一种新的杂波模型参数估计算法。该算法将一个四维非线性最优化问题转化为一个一维线性最优化问题,从而降低了计算量和所需样本数,提高了估计性能。最后,进行了仿真实验,验证了结论。Under high resolution radar, it is more suitable for the generalized gamma probability density function（ GГ- pdf） that characterizes the speckle component and modulation component of the clutter. Firstly this paper sets up the GC-pdf model and deduces its statistical properties, which shows the wide applicable range for the clutters models. Then this paper analyzes the method of least mean squares （LMS） parameter estimation. Furthermore, a new model parameter estimation algorithm is put forward through the parameter decoupling and fully using the information contained in the clutter sequences. This novel estimation method translate a four dimension nonlinear optimization problem to a one dimension linear optimization problem, so it can not only improve the estimation performance, but also depress the calculation quantity and the necessarily clutter sample numbers. Finally, the simulation is processed and the results prove the conclusion proposed in this paper.

关 键 词：广义复合杂波分布 建模 参数解耦 参数估计 

分 类 号：TN955[电子电信—信号与信息处理]