FlowPic: A Generic Representation for Encrypted Traffic Classification and Applications Identification

Shapira, Tal; Shavitt, Yuval

doi:10.1109/tnsm.2021.3071441

Cited by 96 publications

(42 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The works that utilize deep-learning span over multiple scopes and domains, with works that focus on the classification of the operating system, browser, and application levels [1]; mobile app identification [2], [3]; and even webpage fingerprinting [4]. Several works converted the network flow into an image to harness image processing techniques and equivalent Deep-Learning (DL) architectures [2], [11], [17], [18], [20], [21]. Newer works incorporated the Ordinary Differential Equation Network (ODENet) within a DL architecture to classify uni-directional network flows [22].…”

Section: Related Workmentioning

confidence: 99%

“…Along with detection or classification methods, the data in use have great importance as well. Some publicly available datasets such as ISCXVPN2016 are widely used in the literature in works such as [11]- [13], [16], [22], [23], [31]. Where the works adopt the dataset for the purpose of classifying a network flow according to its encapsulation type (tunneled via VPN or not), traffic type (e.g., video/audio/chat), and application (e.g., Skype/Netflix).…”

Section: Related Workmentioning

confidence: 99%

“…Due to the lack of a common shared framework for internet traffic classification, the above tasks are difficult and tiresome. For example, in the application network traffic classification task [11]- [15], researchers would like to compare their features and models using the same framework, especially in systems using the same subset of features (e.g., 784 payload bytes of the flow) [12], [13], [16], [17] and minor changes in the models such as a similar CNN architecture [11], [18].…”

Section: Introductionmentioning

confidence: 99%

“…and an encapsulation label (non-VPN/VPN). Some works that adopted this dataset can be found in [11]- [13], [16], [22], [23], [32].…”

mentioning

confidence: 99%

“…Note that the data contains extensive noise, such as unrelated BlueStacks traffic, Dropbox traffic, and background traffic. Works such as [11] extracted only the relevant flows from each PCAP file, whether the audio flow of Skype or the video flow of Netflix. Many works dropped browser traffic files due to lack of sufficient accurate labeling, and p2p traffic such as BitTorrent due to lack of counterparts in non-encapsulated traffic.…”

mentioning

confidence: 99%

See 4 more Smart Citations

Open-Source Framework for Encrypted Internet and Malicious Traffic Classification

Bader¹,

Lichy²,

Dvir³

et al. 2022

Preprint

View full text Add to dashboard Cite

Internet traffic classification plays a key role in network visibility, Quality of Services (QoS), intrusion detection, Quality of Experience (QoE) and traffic-trend analyses. In order to improve privacy, integrity, confidentiality, and protocol obfuscation, the current traffic is based on encryption protocols, e.g., SSL/TLS. With the increased use of Machine-Learning (ML) and Deep-Learning (DL) models in the literature, comparison between different models and methods has become cumbersome and difficult due to a lack of a standardized framework. In this paper, we propose an open-source framework, named OSF-EIMTC, which can provide the full pipeline of the learning process. From the well-known datasets to extracting new and well-known features, it provides implementations of well-known ML and DL models (from the traffic classification literature) as well as evaluations. Such a framework can facilitate research in traffic classification domains, so that it will be more repeatable, reproducible, easier to execute, and will allow a more accurate comparison of well-known and novel features and models. As part of our framework evaluation, we demonstrate a variety of cases where the framework can be of use, utilizing multiple datasets, models, and feature sets. We show analyses of publicly available datasets and invite the community to participate in our open challenges using the OSF-EIMTC.

show abstract