一种基于指纹特征的加密网页识别方法  

An Encrypted Webpage Identification Method Based on Fingerprint Features

作　　者：陈伟 翟江涛 王勤凡 CHEN Wei;ZHAI Jiangtao;WANG Qinfan(School of Electronics&Information Engineering,Nanjing University of Information Science&Technology,Nanjing 210044)

机构地区：[1]南京信息工程大学电子与信息工程学院,南京210044

出　　处：《计算机与数字工程》2025年第2期460-464,共5页Computer & Digital Engineering

基　　金：国家自然科学基金项目(编号:61931004,62072250);南京信息工程大学人才启动基金项目(编号:2020r061)资助。

摘　　要：加密网页识别是通过对加密网络流量分析得到用户的浏览行为,该技术可以在保护用户隐私和数据的前提下,对犯罪活动进行甄别。然而当前加密网页识别方法很难高精度识别同一网站下的网页应用。因此提出了一种新的加密网页指纹识别方法,所提方法包含指纹特征提取与比对两部分。在指纹特征提取部分,论文方法提取数据包序列长度、时间信息和数据流方向。在指纹比对部分,采用随机森林方法来区分不同网页。对比实验结果表明,论文所提出的网页指纹方法在精度上达到93.5%,与典型方法对比提升了1.5%,降低了时间复杂度,同时稳定性方面优于现有的方法。Encrypted web page identification is to obtain the user's browsing behavior by analyzing encrypted network traffic.This technology can screen criminal activities under the premise of protecting user privacy and data.However,the current encrypted web page identification methods are difficult to identify web applications under the same website with high accuracy.This paper pro⁃poses a new method for fingerprint identification of encrypted web pages.The proposed method includes two parts,which are finger⁃print feature extraction and comparison.In the fingerprint feature extraction part,the method in this paper extracts the data packet sequence length,time information and data flow direction.In the fingerprint comparison part,the random forest method is used to distinguish different web pages.The comparative experimental results show that the web fingerprint method proposed in this paper achieves 93.5%accuracy,which is 1.5%higher than the typical method,reduces the time complexity,and is better than the exist⁃ing methods in terms of stability.

关 键 词：加密流量 机器学习 网页指纹 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TP393.06[自动化与计算机技术—控制科学与工程]