物联网用户目标节点入侵实时定位仿真

摘要

This article puts forward a real-time location method for target node intrusion of RSSI centroid.Firstly,the weighted feature selection and information entropy are utilized to found feature set and to reduce feature dimension,thus evaluated redundancy and correlation of target node and updated feature subset and weight vector of node,and then RSSI value of object node is calculated by radio signal strength and transformed into centroid algorithm weight.Finally,the target node location to be measured is estimated,and intrusion real-time location of target node of internet of things user is completed.The experimental results demonstrate that the method above improves real-time location accuracy of target node and reduces deviation.%对物联网用户目标节点入侵进行定位,能够有效提高物联网用户信息安全性.对物联网用户入侵节点的实时定位,需要先获得目标节点多维特征子集,对其权值向量更新,完成用户目标节点入侵定位.传统方法首先选取低参考线节点为锚节点,进而计算节点定位误差阈值,但忽略了节点特征子集权值向量的更新,导致定位精度低.提出RSSI质心目标节点入侵实时定位方法.利用加权特征选择、信息熵建立特征集,降低特征维度,评估目标节点的冗余性与相关性,对节点的特征子集与权值向量更新.目标节点入侵实时定位,通过无线信号强度对目标节点RSSI值计算,转换为质心算法权值,对待测的目标节点位置估计,完成物联网用户目标节点入侵实时定位.实验结果表明,上述方法提高了目标节点实时定位精度,降低了误差.