Optimised Multi-stage TCP Traffic Classifier Based on Packet Size Distributions

Wang, Xiaoming; Parish, David J.

doi:10.1109/ctrq.2010.24

Cited by 18 publications

(10 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They achieved up to 91% accuracy for a group of protocols such as HTTP, Post Office Protocol 3 (POP3) and Simple Mail Transfer Protocol (SMTP). In a similar work, Wang et al, [45] have considered PDF of the packet size. Their scheme was able to identify a broader range of protocols including file transfer protocol (FTP), Internet Message Access Protocol (IMAP), SSH, and TELNET with accuracy up to 87%.…”

Section: Related Workmentioning

confidence: 99%

Deep packet: a novel approach for encrypted traffic classification using deep learning

et al. 2019

View full text Add to dashboard Cite

Internet traffic classification has become more important with rapid growth of current Internet network and online applications. There have been numerous studies on this topic which have led to many different approaches. Most of these approaches use predefined features extracted by an expert in order to classify network traffic. In contrast, in this study, we propose a deep learning based approach which integrates both feature extraction and classification phases into one system. Our proposed scheme, called "Deep Packet," can handle both traffic characterization in which the network traffic is categorized into major classes (e.g., FTP and P2P) and application identification in which end-user applications (e.g., BitTorrent and Skype) identification is desired. Contrary to most of the current methods, Deep Packet can identify encrypted traffic and also distinguishes between VPN and non-VPN network traffic. After an initial pre-processing phase on data, packets are fed into Deep Packet framework that embeds stacked autoencoder and convolution neural network in order to classify network traffic. Deep packet with CNN as its classification model achieved recall of 0.98 in application identification task and 0.94 in traffic categorization task. To the best of our knowledge, Deep Packet outperforms all of the proposed classification methods on UNB ISCX VPN-nonVPN dataset.

show abstract

Section: Related Workmentioning

confidence: 99%

Deep packet: a novel approach for encrypted traffic classification using deep learning

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Since researchers are more concerned with the technical significance, we categorize most approaches relying on statistical derived terms (heuristics, behavioral, profiling and characterization) under the simple statistical group, at the technique level in Figure . This group includes basic statistical techniques and extended ones such as heuristics and profiles .…”

Section: A Multilateral Taxonomy Of Traffic Classification Methodsmentioning

confidence: 99%

“…Basic statistical techniques rely on general traffic attributes and simple statistical properties to identify applications. For instance, using probability density functions of packet sizes and inter‐arrival times, standard application protocols can be identified with more than 87% overall accuracy. Statistical methods can be extended by constructing sets of experimentally validated rules, such as heuristics or profiles, to describe traffic attributes for specific applications.…”

Section: A Multilateral Taxonomy Of Traffic Classification Methodsmentioning

confidence: 99%

A multilevel taxonomy and requirements for an optimal traffic‐classification model

2014

View full text Add to dashboard Cite

Identifying Internet traffic applications is essential for network security and management. The steady emergence of new Internet applications, together with the use of encryption and obfuscation techniques, ensures that traffic classification remains a hot research topic. Much research has been devoted to this topic by the research community in the last decade. However, an optimal traffic classification model has yet to be defined. Many techniques and formats have been described, with the current literature therefore lacking appropriate benchmarks expressed in a consistent terminology. Moreover, existing surveys are outdated and do not include many recent advances in the field. In this article, we present a systematic multilevel taxonomy that covers a broad range of existing and recently proposed methods, together with examples of vendor classification techniques. Our taxonomy assists in defining a consistent terminology. It could be useful in future benchmarking contexts by characterizing and comparing methods at three different levels. From this perspective, we describe key features and provide design hints for future classification models, while emphasizing the main requirements for promoting future research efforts. To motivate researchers and other interested parties, we collect and share data captured from real traffic, using two models to protect data privacy.

show abstract

“…To avoid the complete reliance on port numbers, many industry products make use of packet payload information. Sen et al [19] have presented a technique for classifying the P2P application traffic by utilizing the application level signatures. Moore and Papa-giannaki [20] have used a combination of port and payload based methods to classify the network applications.…”

Section: Payload-based Segregation Of Ip Trafficmentioning

confidence: 99%

Segregation of IoT Traffic with Machine Learning Techniques

Khedkar¹

2021

TURCOMAT

View full text Add to dashboard Cite

The Internet of Things (IoT) is emerging as a new infrastructure of 21st century. With the advent of cloud computing and evolution of IoT, the classification of traffic over IoT networks has attained significance importance due to rapid growth of users and devices. It is need of the hour to isolate the benign traffic from the malevolent traffic and to channelise the normal traffic to the intended destination to suffice the QoS requirements of the IoT users. A proficient classification mechanism in IoT environment should be capable enough to classify the heavy traffic in a fast manner, to deflect the malevolent traffic on time and to transmit the benign traffic to the designated nodes for serving the needs of the users. In this manuscript, machine learning and deep neural networks-based approaches are proposed for segregating the IoT traffic which eventually enhances the throughput of IoT networks and reduces the congestion over IoT channels. This paper also provides insights into the future research endeavors to channelise the normal traffic and to handle the malicious traffic

show abstract

Optimised Multi-stage TCP Traffic Classifier Based on Packet Size Distributions

Cited by 18 publications

References 14 publications

Deep packet: a novel approach for encrypted traffic classification using deep learning

Deep packet: a novel approach for encrypted traffic classification using deep learning

A multilevel taxonomy and requirements for an optimal traffic‐classification model

Segregation of IoT Traffic with Machine Learning Techniques

Contact Info

Product

Resources

About