基于时空压缩的无线传感器网络入侵检测  

Wireless Sensor Network Intrusion Detection based on Temporal-Spatial Compression

作　　者：康亚楠 张怀相[1] KANG Ya-nan;ZHANG Huai-xiang(School of Computer,Hangzhou Dianzi University,Hangzhou Zhejiang 310018,China)

机构地区：[1]杭州电子科技大学计算机学院,浙江杭州310018

出　　处：《通信技术》2018年第4期901-906,共6页Communications Technology

基　　金：国家科技支撑计划资助项目(No.2014BAF07B01)~~

摘　　要：针对无线传感器网络中的入侵检测问题,为将网络中的资源消耗保持在最低限度的同时实现高检测率和低误报率,提出了一种适用于无线传感器网络的基于时空压缩的神经自编码模型,改进了传统的节点感知数据处理方法,达到了实时提取数据特征的效果。实验仿真使用NSL-KDD数据集,与支持向量机、主成分分析算法进行了对比实验。结果表明,降维至16维时,该模型的入侵检测正确率和F1值分布分别提高了7.84%和15.3%。Aiming at the problem of intrusion detection in wireless sensor networks,in order to keep the resource consumption in the network to a minimum while achieving high detection rate and low false alarm rate,a neural self-encoding model based on temporal-spatial compression for wireless sensor networks is proposed.It improves the traditional node-aware data processing method and achieves the effect of extracting data features in real time.The experimental simulation uses the NSL-KDD data set and compares it with support vector machines and principal component analysis algorithms.The results indicate that the accuracy of the intrusion detection and the distribution of F1 values increased by 7.84%and 15.3%respectively,when the dimension was reduced to 16 dimensions.

关 键 词：无线传感器网络 入侵检测 特征提取 神经自编码 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]