室内Wi-Fi网络下的被动式运动目标检测算法

Indoor Wi-Fi based passive moving target detection approach

传统室内Wi-Fi网络下的被动式运动目标检测方法只提取Wi-Fi信号的均值、方差等粗粒度统计信息,导致系统检测率低。实现被动式运动目标检测的关键是捕获目标对无线链路的影响。探讨了表征原始信号整体分布的方法,构建一种新的相干直方图,并提出基于相干直方图的被动式运动目标检测算法。为解决追踪过程中的位置漂移问题,利用艾伦时间逻辑建立监测区域中不同子区域间的物理逻辑转移关系,对追踪结果进行实时校正。实验结果表明,相较于经典的被动检测技术,该方法基于相干直方图的被动式运动目标检测算法性能更优,综合指标F 1-measure提高近5%。

Some traditional Wi-Fi indoor passive moving target detection systems only extract the coarse-grained statistical information such as the mean value and variance,which lead to low detection accuracy.The key to the realization of passive mo-ving target detection is to characterize the influence of the target on wireless links.This paper explored the method of characte-rizing the distribution of the signal as a whole and proposed a novel coherence histogram.At the same time,this paper proposed a coherence histogram based on passive moving target detection method.In order to solve the position drift problem in the tracking process,it used the Allen time logic to establish the transfer relationship between the different sub-areas in the monitoring area,and it used the relationship to correct the initial result of the position.Compared with classical passive people detection methods,the Wi-Fi indoor passive moving target detection approach based on coherence histogram has better performance.The comprehensive indicator F 1-measure is improved by nearly 5%.