Research on Encrypted Traffic Classification Method Based on Improved Convolutional Neural Network

Zhibo Zhai ； Hui Wang ； Luming Bai

《International Core Journal of Engineering》 6卷6期 (2020/06) Pp. 294-303

https://doi.org/10.6919/ICJE.202006_6(6).0044

摘要

Network traffic classification is of great significance to the daily management and maintenance of the network. With the rapid development of encryption techniques for network traffic, traditional traffic classification methods have reduced the accuracy due to the low identification rate of encrypted traffic. Aiming at this problem, a deep learning traffic classification method based on improved convolutional neural network is proposed. By optimizing and adjusting the network structure, network parameters, and cost function of the convolutional neural network, the classification accuracy of encrypted traffic is improved. By comparing and analyzing the existing traffic classification method on public data sets, the experimental results show that the proposed method effectively improves the classification accuracy of encrypted traffic by 10%.

關鍵字

Traffic classificationd ； Encrypted traffic ； Convolutional neural network ； Deep learning ； Classification accuracy

參考文獻

Yamansavascilar B, Guvensan MA, Yavuz A G, et al. Application Identification Via Network Traffic Classification[C]// 2017 Inter-national Conference on Computing, Networking and Communications (ICNC), 2017: 843-848.

Liu Y, Li W, Li YC. Network Traffic Classification Using K-means Clustering[C]// Second International Multi- Symposiums on Computer and Computational Sciences (IMSCCS), 2007: 360-365.

Tong D, Qu Y R, Prasanna V K. Accelerating Decision Tree Based Traffic Classification on FPGA and Multicore Platforms[J]. IEEE Transactions on Parallel and Distributed Systems, 2017: 3046-3059.

Chen Z, Liu Z, Peng L, et al. A Novel Semi-supervised Learning Method for Internet Application Identification[J]. Soft Computing, 2017, 21(8): 1963-1975.

Munther A, Rozmie R, Mosleh M, et al. A Preliminary Performance Evaluation of K-means, KNN and EM Unsupervised Machine Learning Methods for Network Flow Classification[J]. International Journal of Electrical and Computer Engineering, 2016, 6(2): 778-784.