基于信息熵的多无源传感器数据关联  

Data association for multi-passive-sensor based on information entropy

作　　者：曹乐[1] 王朝英[1] 孔云波[1] 刘玉军[1] 鹿传国 

机构地区：[1]空军工程大学信息与导航学院 [2]95806部队

出　　处：《传感器与微系统》2015年第11期33-37,共5页Transducer and Microsystem Technologies

基　　金：陕西省自然科学基金资助项目(2011JM8023)

摘　　要：针对多维分配模型在构造关联代价函数时,直接利用极大似然估计值代替目标的真实位置信息,未考虑极大似然估计所引入的随机误差问题,提出一种基于信息熵的多无源传感器数据关联算法。考虑到量测后验概率密度函数与伪量测概率密度函数之间的差异性,分别利用相对熵和Renyi熵量化该差异构造关联代价函数以增强模型的完备性。仿真实验结果表明:该算法有效地提高了关联正确率,具有较好的关联性能。Aiming at problem that multi-dimensional assignment model which uses maximum likelihood estimation as true target position has ignored the random errors when constructing the cost function,a data association algorithm is proposed based on information entropy. To improve the completeness of the multi-dimensional assignment model,relative entropy and Renyi entropy are used to quantify the difference between the probability density function of pseudo measurements and the most posterior probability density function to construct the association cost. The simulation results show that the proposed algorithm can improve the correctness and achieve better performance as well.

关 键 词：无源传感器 数据关联 代价函数 相对熵 RENYI熵 

分 类 号：TN953[电子电信—信号与信息处理]