Time Series Analysis for Encrypted Traffic Classification: A Deep Learning Approach
Time Series Analysis for Encrypted Traffic Classification: A Deep Learning Approach
复制标题
加密流量分类的时间序列分析：深度学习方法
DOI：
--
复制
发表时间：
2018
期刊：
影响因子：
--
通讯作者：
E. Dutkiewicz
中科院分区：
文献类型：
--
作者：
Ly Vu;H. V. Thuy;Quang Uy Nguyen;T. N. Ngọc;Diep N. Nguyen;D. Hoang;E. Dutkiewicz
我们开发了一种新的时间序列特征提取技术来解决加密流量/应用分类问题。提出的方法包括两个主要步骤。首先，我们提出了一种特征工程技术，通过分析接收报文的时间序列来提取加密网络流量行为的重要属性。在第二步中，我们开发了一种基于深度学习的技术来利用加密网络应用程序的时间序列数据样本之间的相关性。为了评估所提方法在加密流量分类问题上的效率，我们在一个原始网络流量数据集VPN-non VPN上进行了密集的实验，使用了三个传统的分类器度量：精确度、召回率和F1得分。实验结果表明，该方法在准确率和计算效率方面都能显著提高加密应用流量识别的性能。
We develop a novel time series feature extraction technique to address the encrypted traffic/application classification problem. The proposed method consists of two main steps. First, we propose a feature engineering technique to extract significant attributes of the encrypted network traffic behavior by analyzing the time series of receiving packets. In the second step, we develop a deep learning-based technique to exploit the correlation of time series data samples of the encrypted network applications. To evaluate the efficiency of the proposed solution on the encrypted traffic classification problem, we carry out intensive experiments on a raw network traffic dataset, namely VPN-nonVPN, with three conventional classifier metrics including Precision, Recall, and F1 score. The experimental results demonstrate that our proposed approach can significantly improve the performance in identifying encrypted application traffic in terms of accuracy and computation efficiency.