摘要
为了优化主流分布式检测系统判决性能,提高节点能量利用率,提出了节点多门限机制.通过在常规的并行分布式检测中加入节点多门限机制,在每一个传感器节点都设置有2个门限值,当本地传感器采集的信息量在2个门限值中间时,该节点将处于休眠状态,不发送判决;反之就会有高信息量的传感器节点传输它们的判决到融合中心.实验结果表明:基于节点多门限机制的传感网络,与传统机制相比具有较高的差错指数,实现了降低融合中心的差错概率,从而提高系统的判决性能.
In the current large-scale distributed detection,the sensor nodes transmit binary decisions to the fusion center （FC）. The fusion center also consider the unreliable decision, thus cause the fusion center error probability is increased. Based on more-threshold-based scheme. Set two thresholds at each sensor node,if sensor nodes has less-informative, that will make the nodes in a dormant state, do not sent the decisions,whereas the more informative sensor nodes transmit their decisions to the FC. The experimental results illustrate that more-threshold-based scheme have a higher error exponent, can lower the probability of error at FC,and improve the judgment performance of system.
出处
《东北师大学报（自然科学版）》
CAS
CSCD
北大核心
2013年第1期60-64,共5页
Journal of Northeast Normal University(Natural Science Edition)
基金
国家自然科学基金资助项目(61262040)
云南省应用基础研究计划项目(KKSY201203062)
关键词
判决性能
多门限机制
低能耗
休眠
差错概率
judgment performance
more-threshold-based
low power
dormant
probability of error
