基于相似度计算的物联网传输流丢包节点检测  

Packet Loss Nodes Detection of Internet of Things Transport Stream Based on Similarity Calculation

作　　者：童星[1] TONG Xing(Qingdao University of Technology,Linyi Shandong 273400,China)

机构地区：[1]青岛理工大学,山东临沂273400

出　　处：《计算机仿真》2022年第5期388-392,共5页Computer Simulation

摘　　要：针对传统丢包节点检测方法存在的检测效率低、丢包节点定位精准度差、节点转发率低的问题,设计一种基于相似度计算的物联网传输流丢包节点检测方法。首先构建传感器节点分布模型,并运用二元有向图对其描述,然后根据传输关联信息熵结果创建传输任务信道分布模型;在此基础上,根据丢包节点检测的基本原理对未知分类的节点实施分类检测,并且计算出每个分类的最大似然值,最后对节点进行感测向量检测处理,并将跨度和实际节点之间的相似度作为对应的判定标准,完成丢包节点检测。仿真结果表明:与传统检测方法相比,该方法检测过程效率更高,且通过相似度计算提高了丢包节点定位的精准度,确保了较高的节点转发率,能够很好地适用于对物联网传输任务的检测。In this paper,a method to detect the packet loss node in IoT transmission stream based on similarity calculation was designed.Firstly,the model of sensor node distribution was constructed,and then the model was described by binary directed graph.Secondly,the channel distribution model of transmission task was built by the transmission correlation information entropy.According to the basic principle of packet loss node detection,the nodes of unknown classification were detected,and the maximum likelihood value of each classification was calculated.Finally,the sensing vector of node was detected,and then the similarity between span and actual node was taken as the criterion to complete the detection for packet loss node.Following conclusions can be drawn from simulation results:compared with the traditional method,the proposed method is more efficient during the detection;Meanwhile,this method improves the location accuracy of lost packet node,thus ensuring high node forwarding rate;This method is very suitable for the detection of Internet of things transmission tasks.

关 键 词：物联网 相似度计算 数据传输 丢包 节点检测 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]