基于随机时空联合维度校准机制的WSN恶意节点定位算法

WSN Malicious Node Location Algorithm Based on Random Spatiotemporal Joint Dimension Calibration Mechanism

为提高无线传感网对恶意节点的查证识别能力,提出一种基于随机时空联合维度校准机制的WSN恶意节点定位算法.首先,采取一一映射机制,按照空间和时间维度进行时空建模,以提高网络对恶意行为的敏感程度.根据归一化后的节点均值,采取窗口方案对节点数据予以截断处理,以增强恶意节点的查证效果.随后,结合对角化时空矩阵协方差,采取特征值匹配模型,利用特征值的平均分布对恶意行为进行二次匹配,从而提高对疑似行为的过滤质量.仿真实验表明,与当前无线传感网恶意节点定位领域常用的DAR-UI算法及SC-RFL算法相比,所提算法具有更高的网络恶意节点识别频次和更短的首次恶意行为识别时间.

In order to improve the verification and identification ability of wireless sensor networks to malicious nodes,a WSN malicious node location algorithm based on random spatio-temporal joint dimension calibration mechanism is proposed.Firstly,a one-to-one mapping mechanism is adopted to conduct spatio-temporal modeling according to the spatial and temporal dimensions,so as to improve the sensitivity of the network to malicious behavior.According to the normalized node mean,a window scheme is adopted to truncate the node data to enhance the verification effect of malicious nodes.Then,combined with the diagonalized spatio-temporal matrix covariance,the eigenvalue matching model is adopted,and the average distribution of eigenvalues is used for secondary matching of malicious behavior,so as to improve the filtering quality of suspected behavior.Simulation results show that compared with darui algorithm and scrfl algorithm commonly used in the field of malicious node location in wireless sensor networks,the proposed algorithm has higher network malicious node identification frequency and shorter first malicious behavior identification time.