FLOWGAN:Unbalanced Network Encrypted Traffic Identification Method Based on GAN

Wang, ZiXuan; Wang, Pan; Zhou, Xiaokang; Li, Shuhang; Zhang, Moxuan

doi:10.1109/ispa-bdcloud-sustaincom-socialcom48970.2019.00141

Cited by 29 publications

(30 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One of the major challenges in encrypted traffic classification is class imbalance, given that the majority of network traffic is regular and unencrypted traffic. To this end, Wang et al [77] proposed FlowGAN, a method that uses GANs to generate synthetic traffic data for classes that suffer from low sample counts. They then used an MLP classifier to evaluate the effectiveness of their method, finding that tackling the class imbalance problem in this manner can indeed increase the performance traffic classifiers.…”

Section: Network Analysismentioning

confidence: 99%

“…Li et al [78] proposed FlowGAN (not to be confused with the FlowGAN proposed in [77]), a novel dynamic traffic camouflaging method to mitigate traffic analysis attacks and circumvent censorship. The idea behind this method is to use a GAN to learn features of permitted network flow (the target) and morph on-going censored traffic flows (the source) based on these features, in such a way that the morphed traffic is indistinguishable from the real flow.…”

Section: Network Analysismentioning

confidence: 99%

See 1 more Smart Citation

Generative Adversarial Networks (GANs) in networking: A comprehensive survey & evaluation

et al. 2021

View full text Add to dashboard Cite

show abstract

Section: Network Analysismentioning

confidence: 99%

Section: Network Analysismentioning

confidence: 99%

Generative Adversarial Networks (GANs) in networking: A comprehensive survey & evaluation

et al. 2021

View full text Add to dashboard Cite

show abstract

“…No details are given of either the structure of the GAN or how the experiments were evaluated, making it impossible to reproduce the proposed solution or measure the synthetic data quality. Another similar work [23] used GANs and the "ISCX VPN nonVPN" traffic dataset [24] but without proposing any evaluation method to contrast the obtained results.…”

Section: Related Workmentioning

confidence: 99%

“…• Discriminator. Architecture: [4,380,800,600,177,23,1], output filtering: False, noise in input (real and fakes): N(0,0.02), noise in fakes: N(0,0), ratio label change: 0, batch normalization: True, regularization 0.02, dropout:0.1, learning rate RMS-Prop:0.001.…”

Section: Standard Ganmentioning

confidence: 99%

Synthetic flow-based cryptomining attack generation through Generative Adversarial Networks

Mozo,

González-Prieto,

Pastor

et al. 2021

Preprint

View full text Add to dashboard Cite

Due to the growing rise of cyber attacks in the Internet, flow-based data sets are crucial to increase the performance of the Machine Learning (ML) components that run in network-based intrusion detection systems (IDS). To overcome the existing network traffic data shortage in attack analysis, recent works propose Generative Adversarial Networks (GANs) for synthetic flow-based network traffic generation. Data privacy is appearing more and more as a strong requirement when processing such network data, which suggests to find solutions where synthetic data can fully replace real data. Because of the ill-convergence of the GAN training, none of the existing solutions can generate high-quality fully synthetic data that can totally substitute real data in the training of IDS ML components. Therefore, they mix real with synthetic data, which acts only as data augmentation components, leading to privacy breaches as real data is used.In sharp contrast, in this work we propose a novel deterministic way to measure the quality of the synthetic data produced by a GAN both with respect to the real data and to its performance when used for ML tasks. As a byproduct, we present a heuristic that uses these metrics for selecting the best performing generator during GAN training, leading to a stopping criterion. An additional heuristic is proposed to select the best performing GANs when different types of synthetic data are to be used in the same ML task. We demonstrate the adequacy of our proposal by generating synthetic cryptomining attack traffic and normal traffic flow-based data using an enhanced version of a Wasserstein GAN. We show that the generated synthetic network traffic can completely replace real data when training a ML-based cryptomining detector, obtaining similar performance and avoiding privacy violations, since real data is not used in the training of the ML-based detector.

show abstract

“…One of the major challenges in encrypted traffic classification is class imbalance, given that the majority of network traffic is regular and unencrypted traffic. In this context, Wang et al [77] proposed FlowGAN, a method that uses GANs to generate synthetic traffic data for classes that suffer from low sample counts. They then used an MLP classifier to evaluate the effectiveness of their method, finding that tackling the class imbalance problem in this manner can indeed increase the performance traffic classifiers.…”

Section: Network Analysismentioning

confidence: 99%

Generative Adversarial Networks (GANs) in Networking: A Comprehensive Survey & Evaluation

Navidan,

Moshiri,

Nabati

et al. 2021

Preprint

View full text Add to dashboard Cite

Despite the recency of their conception, Generative Adversarial Networks (GANs) constitute an extensively-researched machine learning sub-field for the creation of synthetic data through deep generative modeling. GANs have consequently been applied in a number of domains, most notably computer vision, in which they are typically used to generate or transform synthetic images. Given their relative ease of use, it is therefore natural that researchers in the field of networking (which has seen extensive application of deep learning methods) should take an interest in GAN-based approaches. The need for a comprehensive survey of such activity is therefore urgent. In this paper, we demonstrate how this branch of machine learning can benefit multiple aspects of computer and communication networks, including mobile networks, network analysis, internet of things, physical layer, and cybersecurity. In doing so, we shall provide a novel evaluation framework for comparing the performance of different models in non-image applications, applying this to a number of reference network datasets.

show abstract

FLOWGAN:Unbalanced Network Encrypted Traffic Identification Method Based on GAN

Cited by 29 publications

References 15 publications

Generative Adversarial Networks (GANs) in networking: A comprehensive survey & evaluation

Generative Adversarial Networks (GANs) in networking: A comprehensive survey & evaluation

Synthetic flow-based cryptomining attack generation through Generative Adversarial Networks

Generative Adversarial Networks (GANs) in Networking: A Comprehensive Survey & Evaluation

Contact Info

Product

Resources

About