基于图注意力机制的加密移动应用同质流量识别方法
网络出版日期: 2024-07-23
基金资助
国家自然科学基金联合基金项目(U22B2025)
版权
Homogeneous traffic identification method for encrypted mobile applications based on graph attention mechanism
Online published: 2024-07-23
Supported by
The Joint Funds of the National Natural Science Foundation of China(No. U22B2025)
Copyright
为了满足加密移动应用流量识别与管理的需求,针对加密移动应用中使用相似的第三方库而导致同质化流量的问题,提出了一种基于图注意力机制的加密移动应用同质流量识别方法。该方法首先将突发中的多条流相互连接构造流量相关图,在图节点中嵌入数据包特征;其次使用门控循环单元对流量相关图节点内部特征进行学习,使用图注意力神经网络对每个图节点的邻居节点进行上下文流量特征的重要性建模,增强对同质流量中上下文流量的关注与学习,达到识别同质流量的目的。考虑真实场景下存在背景流量与待识别移动应用流量同时存在的情况,在所构建的流量识别模型中加入图节点随机扰动和一致性正则化损失来增强模型的鲁棒性。基于公开数据集,通过与相关研究成果进行实验比较,突出了所提出的方法在加密移动应用同质化流量识别方面的性能优化效果。
孙泽沛 , 王子豪 , 潘炜 . 基于图注意力机制的加密移动应用同质流量识别方法[J]. 网络空间安全科学学报, 2024 , 2(2) : 97 -106 . DOI: 10.20172/j.issn.2097-3136.240210
For the purpose of traffic identification and management of encrypted mobile applications, a homogeneous traffic identification method for encrypted mobile applications based on graph attention mechanism was proposed to solve the problem of traffic homogeneity caused by the use of similar third-party libraries in encrypted mobile applications. Firstly, the traffic correlation graphs were constructed by multiple streams in traffic bursts, and packet features were embedded in the graph nodes. Then, the gated recurrent units were used to learn internal features of traffic correlation graph nodes. Meanwhile, the graph attention neural network was used to model the importance of the contextual traffic features of neighbor nodes of each graph node, so as to enhance the attention and learning of the contextual traffic in the homogeneous traffic. The proposed method can achieve the purpose of identifying the homogeneous traffic. Considering the coexistence of background traffic and mobile application traffic in the real scenario, the stochastic perturbation of graph nodes and the regularization loss of consistency were introduced to the proposed traffic identification model to enhance the robustness of the model. Experiments were conducted based on the public traffic dataset. The experimental comparison and analysis demonstrat that the proposed method can identify the homogeneous traffic identification of encrypted mobile applications, and provide better performance optimization compared to other approaches.
表 1 EHT-GAT模型的算法Table 1 The algorithm of EHT-GAT |
| 算法:基于图注意力机制的加密移动应用同质流量识别算法 |
| 输入: 邻接矩阵 输出: 预测结果Z While epoch in epochs do for t = 1:T do 生成扰动邻接矩阵 使用GRU对所有图节点分别进行时序特征学习: 对 使用全连接和softmax输出输出所有节点的标签预测张量: end for 使用交叉熵损失函数计算监督分类损失 通过梯度下降更新模型参数: end while 输出预测 |
表 2 EHT-GAT模型详细参数Table 2 Detailed parameters of the EHT-GAT |
| 层数 | 结构 | 输入尺寸 | 模型参数 |
| 1 | LayerNorm | (M, N, 256) | normalized_shape = (N, 256) |
| 2 | GRU | (M, N, 256) | hidden_size = 256 |
| 3 | GRU | (M, N, 256) | hidden_size = 256 |
| 4 | Flatten | (M, N, 256) | start_dim = 1 |
| 5 | DropOut | (M, N×256) | rate = 0.3 |
| 6 | GAT | (M, N×256), (M, M) | head =8, dropout = 0.3 |
| 7 | DropOut | (M, N×256) | rate = 0.3 |
| 8 | Dense | (M, N×256) | out_features = ClassNum |
表 3 在Cross-Platform(安卓)上的实验结果Table 3 The results of the experiment on Cross-Platform (Android) |
| 算法 | 准确率 | 查准率 | 召回率 | F1值 |
| Fs-Net | ||||
| App-Net | ||||
| GraphDApp | ||||
| GCNA | ||||
| FlowPrint | ||||
| EHT-GAT(本文) |
表 4 在Cross -Platform(iOS)上的实验结果Table 4 The results of the experiment on Cross-Platform (iOS) |
| 算法 | 准确率 | 查准率 | 召回率 | F1值 |
| Fs-Net | ||||
| App-Net | ||||
| GraphDApp | ||||
| GCNA | ||||
| FlowPrint | ||||
| EHT-GAT(本文) |
表 5 去除Cross-Platform同质流量的实验结果Table 5 The results on Cross-Platform without homogeneous flows |
| 算法 | 准确率 | 查准率 | 召回率 | F1值 |
| Fs-Net | ||||
| App-Net | ||||
| GraphDApp | ||||
| GCNA | ||||
| FlowPrint | ||||
| EHT-GAT(本文) |
| 1 |
WANG Y,AN J,HUANG W. Using CNN-based representation learning method for malicious traffic identification[C]//2018 IEEE/ACIS 17th International Conference on Computer and Information Science (ICIS),Singapore:IEEE,2018:400-404.
|
| 2 |
LI J,YUN X C,TIAN M,et al. A method of HTTP malicious traffic detection on mobile networks[C]//2019 IEEE Wireless Communications and Networking Conference (WCNC),Marrakesh,Morocco:IEEE,2019:1-8.
|
| 3 |
KARA I,AYDOS M. Cyber Fraud:Detection and analysis of the crypto-ransomware[C]//2020 11th IEEE Annual Ubiquitous Computing,Electronics & Mobile Communication Conference (UEMCON),New York,NY,USA:IEEE,2020:0764-0769.
|
| 4 |
KATARIA S,NAFIS M T. Internet banking fraud detection using deep learning based on decision tree and multilayer perceptron[C]//2019 6th International Conference on Computing for Sustainable Global Development (INDIACom),New Delhi,India:IEEE,2019:1298-1302.
|
| 5 |
WANG P, CHEN X, YE F, et al. A survey of techniques for mobile service encrypted traffic classification using deep learning[J]. IEEE Access, 2019, (7): 54024- 54033.
|
| 6 |
SHEN M, WEI M, ZHU L, et al. Classification of encrypted traffic with second-order markov chains and application attribute bigrams[J]. IEEE Transactions on Information Forensics and Security, 2017, 12 (8): 1830- 1843.
|
| 7 |
TAYLOR V F, SPOLAOR R, CONTI M, et al. Robust smartphone App identification via encrypted network traffic analysis[J]. IEEE Transactions on Information Forensics and Security, 2017, 13 (1): 63- 78.
|
| 8 |
BACKES M,BUGIELl S,DERR E. Reliable third-party library detection in android and its security applications[C]//Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security,2016:356-367.
|
| 9 |
DOROUD H,ALASWAD A,DRESSLER F. Encrypted traffic detection:Beyond the port number era[C]//2022 IEEE 47th Conference on Local Computer Networks (LCN),Edmonton,AB,Canada:IEEE,2022:198-204.
|
| 10 |
YE K, SHEN H, WANG Y, et al. Multi-tier workload consolidations in the cloud: Profiling modeling and optimization[J]. IEEE Trans. Cloud Comput., 2022, 10 (2): 899- 912.
|
| 11 |
EL-MAGHRABY R T,ELAZIM N M A,BAHAA-ELDIN A M. A survey on deep packet inspection[C]//2017 12th International Conference on Computer Engineering and Systems (ICCES),Cairo,Egypt:IEEE,2017:188-197.
|
| 12 |
KHANDAIT P,HUBBALLI N,MAZUMDAR B. Efficient keyword matching for deep packet inspection based network traffic classification[C]//2020 International Conference on COMmunication Systems & NETworkS (COMSNETS),Bengaluru,India:IEEE,2020:567-570.
|
| 13 |
陈子涵, 程光, 徐子恒, 等. 互联网加密流量检测、分类与识别研究综述[J]. 计算机学报, 2023, 46 (5): 1060- 1085.
CHEN Z H, CHENG G, XU Z H, et al. A survey on internet encrypted traffic detection, classification and identification[J]. Chinese Journal of Computers, 2023, 46 (5): 1060- 1085.
|
| 14 |
BARUT O,ZHU R,LUO Y,et al. TLS encrypted application classification using machine learning with flow feature engineering[C]//ICCNS 2020:2020 the 10th International Conference on Communication and Network Security,New York,USA:Association for Computing Machinery,2020:32-41.
|
| 15 |
RAMIRES M,GOMES A S,RITO LIMA S,et al. Network traffic classification using ML:A comparative analysis[C]//2022 17th Iberian Conference on Information Systems and Technologies (CISTI),Madrid,Spain,2022:1-6.
|
| 16 |
WANG W ,ZHU M ,WANG J ,et al. End-to-end encrypted traffic classification with one-dimensional convolution neural networks[C]//2017 IEEE International Conference on Intelligence and Security Informatics (ISI),Beijing,China:IEEE,2017:43-48.
|
| 17 |
HE H Y,GUO Y Z,CHEN X N. PERT:Payload encoding representation from transformer for encrypted traffic classification[C]//2020 ITU Kaleidoscope:Industry-Driven Digital Transformation (ITU K),Ha Noi,Vietnam:IEEE,2020:1-8.
|
| 18 |
REZAEI S, LIU X. Deep learning for encrypted traffic classification: An overview[J]. IEEE Communications Magazine, 2019, 57 (5): 76- 81.
|
| 19 |
TAYLOR V F,SPOLAOR R,CONTI M,et al. Appscanner:Automatic fingerprinting of smartphone Apps from encrypted network traffic[C]//2016 IEEE European Symposium on Security and Privacy (EuroS&P),Saarbruecken,Germany:IEEE,2016:439-454.
|
| 20 |
VAN E T,BORTOLAMEOTTI R,CONTINELLA A,et al. Flowprint:Semi-supervised mobile-app fingerprinting on encrypted network traffic[C]//In Proceedings of the ISOC Network and Distributed Systems Security Symposium (NDSS),2020:1-18.
|
| 21 |
LIU C,HE L,XIONG G,et al. Fs-Net:A flow sequence network for encrypted traffic classification[C]//IEEE INFOCOM 2019-IEEE Conference on Computer Communications,Paris,France:IEEE,2019:1171-1179.
|
| 22 |
WANG X,CHEN S,SU J. App-Net:A hybrid neural network for encrypted mobile traffic classification[C]//IEEE INFOCOM 2020-IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS),Toronto,ON,Canada:IEEE,2020:424-429.
|
| 23 |
崔华俊, 孟国柱, 李玥琦, 等. 基于联邦学习的第三方库流量识别[J]. 信息安全学报, 2023, 8 (3): 128- 145.
CUI H J, MENG G Z, LI Y Q et al. A third-party library traffic identification framework using federated learning[J]. Journal of Cyber Security, 2023, 8 (3): 128- 145.
|
| 24 |
SHEN M, ZHANG J, ZHU L, et al. Accurate decentralized application identification via encrypted traffic analysis using graph neural networks[J]. IEEE Transactions on Information Forensics and Security, 2021, 16, 2367- 2380.
|
| 25 |
SUN B,YANG W,YAN M,et al. An encrypted traffic classification method combining graph convolutional network and autoencoder[C]//2020 IEEE 39th International Performance Computing and Communications Conference (IPCCC),Austin,TX,USA:IEEE,2020:1-8.
|
| 26 |
CHEN Y,ZANG T,ZHANG Y,et al. Rethinking encrypted traffic classification:A multi-attribute associated fingerprint approach[C] //2019 IEEE 27th International Conference on Network Protocols (ICNP),Chicago,IL,USA:IEEE,2019:1-11.
|
| 27 |
MISKOVIC S,LEE G M,LIAO Y,et al. Appprint:Automatic fingerprinting of mobile applications in network traffic[C]//Proceedings of the 16th International Conference on Passive and Active Measurement,PAM 2015,New York City,USA,2015:57-69.
|
| 28 |
VELICKOVIC P,CUCURULLl G,CASANOVA A,et al. Graph attention networks[C]//Proceedings of the 6th International Conference on Learning Representations. Vancouver:ICLR,2018:1-12.
|
| 29 |
WANG Y, YUN X, ZHANG Y, et al. A multi-scale feature attention approach to network traffic classification and its model explanation[J]. IEEE Transactions on Network and Service Management, 2022, 19 (2): 875- 889.
|
| 30 |
REN J,DUBOIS D J,CHOFFNES D. An international view of privacy risks for mobile Apps[EB/OL]. [2023-09-03]. https://recon.meddle.mobi/papers/ cross-market.pdf,2019.
|
/
| 〈 |
|
〉 |