传感技术学报2025，Vol.38Issue(10)：1892-1897,6.DOI:10.3969/j.issn.1004-1699.2025.10.022
无线传感器网络安全时序数据流多层次提取方法
A Multi-Level Extraction Method for Secure Temporal Data Streams in Wireless Sensor Networks
摘要
Abstract
Wireless sensor network data flow is often produced with high dimension and complexity,contains data of a large number of sensor nodes and data of a variety of types of sensors,leading to has certain difficulty for data extracted.Therefore,a multi-level extrac-tion method of security time series data flow in wireless sensor networks is proposed.Wavelet decomposition and reconstruction are per-formed for the original signal flow,the time series data stream is decomposed into several layers,clustering characteristics are retained,the improved competition algorithm is used ot train self-organization process neural networks,the temporal characteristics of the signals are input to the self-organization process neural networks,and rough clustering processing of sequential data stream is performed.The fluctuation Angle and range coefficient of the time series data flow are calculated,and the data flow is divided into abnormal and normal types according to the preset steady fluctuation value,which are stacked to realize multi-level extraction of the time series data flow.The simulation analysis shows that the average silhouette coefficient of the proposed method is above 0.8,and the average DBI index is not less than 0.54,which indicates that the proposed method can effectively and accurately extract the secure time series data stream of wireless sensor network.关键词
无线传感器网络/多层次提取/小波分解/改进竞争算法/安全时序数据流Key words
wireless sensor network/multilevel extraction/wavelet decomposition/improved competition algorithm/secure time series data flow分类
计算机与自动化引用本文复制引用
韩俊芳,任瑞仙,李军红..无线传感器网络安全时序数据流多层次提取方法[J].传感技术学报,2025,38(10):1892-1897,6.基金项目
山西省教育科学十三五规划课题项目(GH-19144) （GH-19144）
山西省教育科学十四五规划课题项目(GH-220825) （GH-220825）
教育部产学合作协同育人项目(220501363101014) （220501363101014）
