基于多维相似度的网络传输通道恶意入侵检测方法

Malicious Intrusion Detection Method of Network Transmission Channel Based on Multi-dimensional Similarity

为解决现有网络传输通道入侵检测方法存在的稳定性较差、误检率较高的问题,提出基于多维相似度的网络传输通道恶意入侵检测方法。利用2个数据样本间的差异程度来判断数据的多维相似度,再根据数据间的相邻数目来判断数据是否存在异常,并完成对异常数据的挖掘;在此基础上,依据用户的行为或是对系统的操作情况,结合判断是否存在恶意入侵的现象完成对网络传输通道中恶意入侵的检测。仿真实验证明,与传统方法相比,该方法检测过程的稳定性较高,且校测结果的精准度也较高,具有较强的实用性。

In order to solve the problems of poor stability and high error-detection rate in the existing network transmission channel intrusion detection methods,a malicious intrusion detection method based on multidimensional similarity is proposed.The difference degree between two data samples is used to judge the multidimensional similarity of data,and then the number of adjacent data samples is used to judge whether there are anomalies in the data,and the mining of abnormal data is completed.On this basis,according to the user’s behavior or the operation of the system,combined with the judgment of whether there is malicious intrusion phenomenon to complete the detection of malicious intrusion in the network transmission channel.The simulation results show that compared with the traditional method,this method has higher stability in the detection process and higher accuracy in the calibration results.