A Survey of Techniques for Mobile Service Encrypted Traffic Classification Using Deep Learning

Abstract

The rapid adoption of mobile devices has dramatically changed the access to various networking services and led to the explosion of mobile service traffic. Mobile service traffic classification has been a crucial task that attracts strong interest in mobile network management and security as well as machine learning communities for past decades. However, with more and more adoptions of encryption over mobile services, it brings a lot of challenges about mobile traffic classification. Although classical machine learning approaches can solve many issues that port and payload-based methods cannot solve, it still has some limitations, such as time-consuming, costly handcrafted features, and frequent features update. With the excellent ability of automatic feature learning, Deep Learning (DL) undoubtedly becomes a highly desirable approach for mobile services traffic classification, especially encrypted traffic. This survey paper looks at emerging research into the application of DL methods to encrypted traffic classification of mobile services and presents a general framework of DL-based mobile encrypted traffic classification. Moreover, we review most of the recent existing work according to dataset selection, model input design, and model architecture. Furthermore, we propose some noteworthy issues and challenges about DL-based mobile services traffic classification.

An Improved Network Traffic Classification Model Based on a Support Vector Machine

“…Deep learning is a characterization learning method based on the data in machine learning [17]. Though the training of deep learning depends on the huge amount of training data heavily, and the hyper-parameters optimization of deep neural networks is difficult [18]. In the above methods, the support vector machine (SVM) algorithm is a kind of supervised learning algorithm based on statistical theory [19].…”

CAPC: Packet-Based Network Service Classifier With Convolutional Autoencoder

“…5 48 after value scaling. For example, the first byte ( , ) in 49 the first packet 1 ( ) is originally 0×69 (105 in decimal), 50 while it becomes 0.41 in Fig. 6 after proportional scaling.…”

“…Fig. 1 shows that two approaches can be categorized 61 depending on the training data type: flow-based and packet-62 based [13]. 63 64 FIGURE 1.…”

Encrypted Traffic Classification Based on Unsupervised Learning in Cellular Radio Access Networks

“…Thus, classification models must be retrained with new data periodically [19]. To overcome this issue, other works propose classifiers based on deep learning, that work directly on input data by automatically distilling structured and complex feature representations at the expense of a higher training complexity and need for larger datasets [14]. In wireless networks, this approach has been considered via variational autoencoder networks [21], convolutional networks [22] or multi-modal classifiers [6] [23].…”

“…Moreover, these methods depend on large quantities of labeled data, which are difficult to obtain. For these reasons, the design of semi-supervised [12] or unsupervised [13] schemes is considered a promising research direction [14]. Nonetheless, in the particular case of mobile networks, both supervised or unsupervised flow-based traffic classification require probes that analyze traffic in the core network.…”