基于随机观测集和信息论的多传感器管理算法  

Sensor Management Based on Random Set Observations and Information Theory

作　　者：罗文涛[1] 许蕴山[1] 向建军[1] 夏海宝[1] 

机构地区：[1]空军工程大学航空航天工程学院,西安710038

出　　处：《现代雷达》2013年第12期38-42,46,共6页Modern Radar

基　　金：空军工程大学工程学院科研创新基金预警机多传感器管理技术研究(XS1101011)

摘　　要：针对传感器网络中传感器对目标的分配与识别问题,研究了自适应地控制多传感器的观测信息。采用一种基于随机观测集的统计结构,模拟被观测特征与目标类型之间的统计关系。通过利用最佳值分配算法,提出了一种计算随机集间分配差异的新算法。仿真结果表明,同基于仿真模拟的自适应方法相比,基于信息的传感器管理算法通过估计信息的价值能实现更准确的分配,并且能减少近3个数量级的计算量。Consider the distribution and identification problem of sensors to targets in the sensors network, the approach of control- ling the observations of these sensors adaptively is discussed in this paper. Also a statistical framework based on random sets is ex- ploited to model the statistical relationship between observed features and object types. In particular, a novel approach for computa- tion of these distances between distributions of random sets is presented by using k-best assignment algorithms. Simulation results show that the information-based sensor management algorithms achieve comparable classification accuracy to adaptive simulation- based approaches that evaluate the value of information, while requiring nearly three orders of magnitude less computation.

关 键 词：多传感器管理 随机观测集 切尔洛夫系数 信息理论 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置]