混合群智感知网络的异常状态检测仿真研究  

Simulation Research on Abnormal State Detection of Hybrid Group Intelligence Network

作　　者：薄文彦[1] BO Wen-yan(School of Computer and Network Engineering,Shanxi Datong University,Datong Shanxi 037009,China)

机构地区：[1]山西大同大学计算机与网络工程学院,山西大同037009

出　　处：《计算机仿真》2020年第11期358-361,共4页Computer Simulation

基　　金：大同市科技计划:大同市科技成果转化信息服务平台建设(2018187)。

摘　　要：采用当前方法检测混合群智感知网络的状态时,检测所用的时间较长,受噪声信号的干扰较大,检测结果与实际结果不符,存在检测效率低、去噪效果差和检测准确率低的问题。提出混合群智感知网络的异常状态检测方法,结合基于聚类的样本选择方法和过滤器模式的特征选择方法从混合群智感知网络中采集数据构建数据集。利用阈值自学习小波算法去除数据集中存在的噪声,在核密度估计理论的基础上估计数据集中数据的概率密度,根据估计结果建立信任函数,计算被检测数据在混合群智感知网络中的信任度,利用信任度判断混合群智感知网络的状态,完成混合群智感知网络的异常状态检测。仿真结果表明,所提方法的检测效率高、去噪效果好、检测准确率高。In current methods of detecting the state of hybrid crowd-sensing network,the detection efficiency and the detection accuracy are low.In this article,a method to detect abnormal state based on hybrid crowd-sensing network was proposed.Combined the sample selection method based on the clustering with the feature selection method based on filter pattern,the data set was constructed by collecting the data from the hybrid crowd-sensing network.The threshold self-learning wavelet algorithm was used to remove the noise in data set.The probability density of data in data set was estimated by the kernel density estimation theory.Based on the estimated results,the trust function was established to calculate the trust degree of the detected data in hybrid crowd-sensing network.The state of hybrid crowd-sensing network was judged by the trust degree.Thus,the detection of abnormal state of hybrid crowd-sensing network was completed.Simulation results show that the proposed method has high detection efficiency,good denoising effect and high detection accuracy.

关 键 词：混合群智感知网络 异常状态 检测方法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]