无线传感器网络恶意信标干扰信号协同过滤方法  

Collaborative Filtering Method for Malicious Beacon Interference Signals in Wireless Sensor Networks

作　　者：刘海峰 徐丽丽 武堂颖 LIU Haifeng;XU Lili;WU Tangying(Big Data College,Jinzhong College of Information,Jinzhong Shanxi 030800,China)

机构地区：[1]晋中信息学院大数据学院,山西晋中030800

出　　处：《传感技术学报》2024年第11期1971-1975,共5页Chinese Journal of Sensors and Actuators

基　　金：山西省高等学校科技创新项目(2023L525);山西省高等学校科技创新项目(2022L666);山西省高等学校教学改革创新项目(J20221444)。

摘　　要：为了保证网络数据传输安全且稳定,提出了无线传感器网络恶意信标干扰信号协同过滤方法。采用短时傅里叶变换小波包法分解接收信号,构建采集完整信号数据的观测环境模型,并采用似然多元分类方法分类网络节点特征矢量,进而明确各个节点的属性。在此基础上,通过区域划分方法对节点特征建模,结合拉普拉斯分布列计算抽样积分安全度,并采用蜜罐模式诱导节点,得到裁决阈值,实现恶意信标干扰信号协同过滤。仿真结果表明,所提方法在数据传输率为224 kbit/s的情况下,过滤率为97%,节点误判率在5%以内,且恶意信标干扰节点收敛效果相对较好。说明恶意信标干扰信号协同过滤方法的恶意信标干扰信号协同过滤效果较佳,可以确保无线传感器网络运行安全顺畅。In order to ensure the security and stability of network data transmission,a collaborative filtering method for malicious beacon interference signals in the wireless sensor network is proposed.The short-time Fourier transform wavelet packet method is used to decompose the received signals,the observation environment model is built for collecting complete signal data,and the likelihood multivariate classification method is used to classify the feature vector of network nodes,so as to clarify the properties of each node.On this basis,by modeling the node features based on region partitioning,and using Laplace distribution column calculation for sampling integration security degrees,a decision threshold is derived.Furthermore,a honeypot mode is adopted to induce the nodes,enabling effective collaborative filtering of malicious beacon interference signals.The simulation results demonstrate that under a data transmission rate of 224 kbit/s,the filtering rate reaches 97%,with a node misjudgment rate below 5%.Moreover,the convergence performance for filtering out malicious beacon interference nodes is relatively good.These results highlight the effectiveness of the collaborative filtering method for malicious beacon interference signals,ensuring the secure and smooth operation of wireless sensor networks.

关 键 词：无线传感器网络 干扰信号 协同过滤 恶意信标节点 观测矩阵 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]