Datanet: Deep Learning Based Encrypted Network Traffic Classification in SDN Home Gateway

Wang, Pan; Ye, Feng; Chen, Xuejiao; Qian, Yi

doi:10.1109/access.2018.2872430

Cited by 168 publications

(82 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Auto encoders are employed [13]- [15], for application identification, where the AEs are utilized in unsupervised fashion to obtain lower representation of the input data. This lower dimensional data is later used as part of a classifier.…”

Section: Related Workmentioning

confidence: 99%

Semi-Supervised Encrypted Traffic Classification With Deep Convolutional Generative Adversarial Networks

Iliyasu

Deng

2020

IEEE Access

View full text Add to dashboard Cite

Network traffic classification serves as a building block for important tasks such as security and quality of service management. The field has been studied for a long time, with many techniques such as classical machine learning and deep learning methods currently available. However, the emergence of stronger encryption protocols has led to the rise of new challenges. One of the challenges is capturing and labeling a large amount of encrypted traffic data especially for training deep learning classifiers, as current techniques rely on deep packet inspection tools (DPI) which perform poorly on encrypted traffic. In this paper, we propose a semi-supervised learning approach using Deep Convolutional Generative Adversarial Network (DCGAN). The basic idea is to utilize the samples generated by DCGAN generators as well as unlabeled data to improve the performance of a classifier trained on a few labeled samples. Thus, alleviating the difficulties associated with large dataset collecting and labeling. To demonstrate the efficacy of our approach, we evaluated our model using a self-collected dataset of the recently established QUIC protocol as well as publicly available ISCX VPN-NonVPN dataset. Our approach is able to achieve 89% and 78% accuracy with a very small number of labeled samples (just 10% of the dataset) on both QUIC and ISCX VPN-NonVPN datasets respectively. INDEX TERMS Deep convolutional generative adversarial network, encrypted traffic classification, semi-supervised learning.

show abstract

Section: Related Workmentioning

confidence: 99%

Semi-Supervised Encrypted Traffic Classification With Deep Convolutional Generative Adversarial Networks

Iliyasu

Deng

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…51 Finally, the normalized data would be saved as ( ) for 52 models to read and train. 53 has been widely applied in packet processing [22], [47], [51]…”

Section: ( ) = ( ) 255mentioning

confidence: 99%

“…highlight the performance of CAPC; these models include 55 DNN, DAE, and 1D CNN, which have been evaluated in 56 several dissertations [22], [47], [48] and gained excellent 57 performance. However, the classification task of the public 58 VPN dataset in the works [22], [47], [48] merely involve 17, 59 15, and 12 applications respectively. To obtain better 60 performance in the more complex scenario (24 applications), 61 the proposed CAPC is compared with the three models 62 63 64 65 66 utilized in those researches.…”

mentioning

confidence: 99%

CAPC: Packet-Based Network Service Classifier With Convolutional Autoencoder

2020

View full text Add to dashboard Cite

The Internet has been evolving from a traditional mechanism to a modern service-oriented architecture, such as quality-of-service (QoS) policies, to meet users' various requirements for high service quality. An instant and effective network traffic classification method is indispensable to identify network services to enforce QoS policies on the corresponding service. Network managers can easily flexibly deploy traffic classification modules and configure the network policies with the help of the emerging softwaredefined networking. However, most existing traffic classification solutions, such as port-based methods or deep packet inspection, cannot handle real-time and encrypted traffic classification. In this research, a Convolutional Autoencoder Packet Classifier (CAPC) has been proposed to immediately classify incoming packets in fine-grained and coarse-grained manners, that is, classifying a service to a single application and a rough genre, respectively. The CAPC is a packet-based deep learning model consisting of a 1D convolutional neural network and an autoencoder, which can handle dynamic-port and encrypted traffic and even cluster similar applications. This classifier is verified on not only the private self-captured traffic but also a public VPN dataset to demonstrate its performance. Moreover, the CAPC classifies different types of service traffic with an accuracy of over 99.9% on the private dataset of 16 services and over 97% on the public dataset of 24 services, thereby outperforming other deep learning classifiers. Experimental results also show other performance metrics, including stability, average precision, and recall and the highest F1-score values of 15 and 18 services on the private and public datasets, respectively.

show abstract

“…Paper [8] proposed a DL-based fast beamforming method, which has the similar performance with traditional iterative algorithms, but its complexity is lower than these iterative algorithms. In the network layer, DL has been applied into network traffic control [23]- [28], predication [29]- [33] and classification [34], [35]. For instance, paper [23] and paper [24] proposed two kinds of DL-based network traffic control methods, respectively.…”

Section: Introductionmentioning

confidence: 99%

Device Type Identification via Network Traffic and Lightweight Convolutional Neural Network for Internet of Things

et al. 2020

View full text Add to dashboard Cite

Device type identification (DTI) is one of the most important techniques for the management of Internet of things (IoT). Recently, deep learning (DL) has been considered as a powerful tools for classification or identification, and some researches have introduced DL into DTI for advanced performance. However, DL-based DTI generally has high computational complexity and large model sizes, which are unsuitable for IoT. Thus, we proposed a lightweight convolutional neural network-based DTI with low computational complexity and few model sizes. In detail, we remove the redundant fully-connected layers and replace common convolution with separable convolution. Simulation results demonstrate that lightweight CNN-based DTI just has less than 5% of computational complexity and model size of CNNbased DTI, though there are less than 0.5% performance gap between them. INDEX TERMS Lightweight convolutional neural network, network traffic, device type identification (DTI), internet of things (IoT).

show abstract

Datanet: Deep Learning Based Encrypted Network Traffic Classification in SDN Home Gateway

Cited by 168 publications

References 33 publications

Semi-Supervised Encrypted Traffic Classification With Deep Convolutional Generative Adversarial Networks

Semi-Supervised Encrypted Traffic Classification With Deep Convolutional Generative Adversarial Networks

CAPC: Packet-Based Network Service Classifier With Convolutional Autoencoder

Device Type Identification via Network Traffic and Lightweight Convolutional Neural Network for Internet of Things

Contact Info

Product

Resources

About