摘要
多传感器目标检测，是近十多年来的一个研究热点。然而运用多传感器实施对目标的检测同时带来了一个新的问题，即如何能由多个传感器所提供的信息实时作出准确的判决？文章证明运用（Ｋ／Ｎ）规则于分布式多传感器目标检测时，存在最佳Ｋ值，使得系统的检测概率高于任一传感器的检测概率，虚警概率低于任一传感器的虚警概率。
Multisensor data fusion can improve the decision and command ability evidently,and is a focus of discussion in recent years.In this paper,we discuss data fusion in distributed multisensor target detection.It is concluded that,in a target detection system with N distributed sensors,if each sensor has the same detection probability(P d i =P d,i=1,2,…,N,P f is the same as P d)and if K(K=(N-3)/2 while N is odd,or K=N/2+1 while N is even)of those sensors decide the target,the system detection probability P d(K/N) is larger than or equal to any of those sensors,and the system false alarm probability P f(K/N) is less than or equal to any of those sensors.
出处
《电子科技》
1997年第4期43-46,共4页
Electronic Science and Technology
关键词
多传感器
目标检测
检测概率
虚警概率
数据融合
multisensor,target detection,detection probability,false alarm probability,data fusion
