中文摘要：

传统的网络入侵检测速度慢、实时性差,且误报率较高。为此,提出一种基于稀疏向量距离的网络入侵数据检测方法。该方法首先对所获得的网络样本数据进行初步分析,采用K-means算法对样本数据包进行量化处理得到该数据流的位置分布集,使用压缩感知的稀疏编码技术处理,得到数据的稀疏表示,然后通过随机投影获取数据集的二值哈希编码可以近似地表示稀疏向量的距离,与设定的阈值进行比较,判断该数据是否为入侵数据。根据这些稀疏向量的距离能够快速而准确地检测到入侵的网络数据。实验结果表明,相对于传统检测算法,本文算法具有速度快、实时性好、误报率低等优点,使入侵检测系统的性能得到了很大提高,充分确保了网络的安全性。

英文摘要：

The traditional network intrusion detection has a slow speed,poor real-time performance and high false alarm rate. Therefore,a method of network intrusion detection based on sparse vector distance was proposed.This method first carries on the preliminary analysis of the network sample data,K-means algorithm is adopted to get the distribution position of the data stream in the quantization of the data packet,using compressed sensing sparse encoding technology processing data sparse representation,and then through the random projection data acquisition two Hash value encoding can be approximately represented the sparse vector distance,compared with the threshold,to judge whether the data for intrusion data. According to the distance of these sparse vectors,the network data can be detected quickly and accurately. The experimental results show that compared with the traditional detection algorithm,this algorithm has fast speed,good real-time performance,low false alarm rate,the performance of the intrusion detection system has been greatly improved,ensuring the security of the network.