基于对数似然估计的目标定位跟踪新方法  

New Object Tracking and Location Method Based on Logarithm Likelihood Estimation

作　　者：刘哲[1] 卫军胡[2] 黄文准[1] 黄世奇[1] 

机构地区：[1]西京学院电子信息工程系,西安710123 [2]西安交通大学电信学院计算机科学与技术系,西安710049

出　　处：《计算机系统应用》2017年第4期141-147,共7页Computer Systems & Applications

基　　金：国防重点预研项目(20157648)

摘　　要：为避免无源定位中的迭代运算,本文针对多站测角无源定位非线性观测方程,提出利用对数似然估计将其进行伪线性化处理,从而实现目标位置的闭式解算.首先,在基于二维测角观测量的前提下,提出利用对数似然估计法将非线性观测方程转化为伪线性观测方程的数学模型,并推导出用于目标定位算法的闭式解.接着,利用"当前"统计机动模型和卡尔曼滤波,实现对目标的精确跟踪定位.并通过仿真实验验证该闭式解的渐近最优性,从而验证文中理论的有效性.In order to avoid the iterative computations in passive location, based on the nonlinear observation equation of multi station passive location, a closed calculation is presented by converting the nonlinear measurement equations into the pseudo-linear equalities, utilizing Logarithm Likelihood Estimation method. First, a mathematical model for the pseudo-linearization of two dimensional nonlinear angle measurement equations is formulated and then the closed form solution for the target location algorithm is derived. Subsequently, we realize precisely tracking and location of objects based on current statistical model. Their theoretical location performances are proved to be able to attain the corresponding Cramér-Rao bound（CRB） and the simulation experiments are conducted to verify the effectiveness of the theoretical analysis in this paper.

关 键 词：似然估计 无源定位 目标跟踪 卡尔曼滤波 

分 类 号：TN95[电子电信—信号与信息处理]