Traffic categorization aims to classify network traffic into major service types. A modern deep neural network based on temporal sequence modeling is proposed for encrypted traffic categorization. The contemporary techniques such as dilated convolution and residual connection are adopted as the basic building block. The raw traffic files are pre-processed to generate 1-dimensional flow byte sequences and are feed into our specially-devised network. The proposed approach outperforms other existing methods greatly on a public traffic dataset.
Lin YAN
State Key Laboratory of Mathematical Engineering and Advanced Computing
Mingyong ZENG
State Key Laboratory of Mathematical Engineering and Advanced Computing
Shuai REN
State Key Laboratory of Mathematical Engineering and Advanced Computing
Zhangkai LUO
Space Engineering University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Lin YAN, Mingyong ZENG, Shuai REN, Zhangkai LUO, "Encrypted Traffic Categorization Based on Flow Byte Sequence Convolution Aggregation Network" in IEICE TRANSACTIONS on Fundamentals,
vol. E104-A, no. 7, pp. 996-999, July 2021, doi: 10.1587/transfun.2020EAL2102.
Abstract: Traffic categorization aims to classify network traffic into major service types. A modern deep neural network based on temporal sequence modeling is proposed for encrypted traffic categorization. The contemporary techniques such as dilated convolution and residual connection are adopted as the basic building block. The raw traffic files are pre-processed to generate 1-dimensional flow byte sequences and are feed into our specially-devised network. The proposed approach outperforms other existing methods greatly on a public traffic dataset.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.2020EAL2102/_p
Copy
@ARTICLE{e104-a_7_996,
author={Lin YAN, Mingyong ZENG, Shuai REN, Zhangkai LUO, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Encrypted Traffic Categorization Based on Flow Byte Sequence Convolution Aggregation Network},
year={2021},
volume={E104-A},
number={7},
pages={996-999},
abstract={Traffic categorization aims to classify network traffic into major service types. A modern deep neural network based on temporal sequence modeling is proposed for encrypted traffic categorization. The contemporary techniques such as dilated convolution and residual connection are adopted as the basic building block. The raw traffic files are pre-processed to generate 1-dimensional flow byte sequences and are feed into our specially-devised network. The proposed approach outperforms other existing methods greatly on a public traffic dataset.},
keywords={},
doi={10.1587/transfun.2020EAL2102},
ISSN={1745-1337},
month={July},}
Copy
TY - JOUR
TI - Encrypted Traffic Categorization Based on Flow Byte Sequence Convolution Aggregation Network
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 996
EP - 999
AU - Lin YAN
AU - Mingyong ZENG
AU - Shuai REN
AU - Zhangkai LUO
PY - 2021
DO - 10.1587/transfun.2020EAL2102
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E104-A
IS - 7
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - July 2021
AB - Traffic categorization aims to classify network traffic into major service types. A modern deep neural network based on temporal sequence modeling is proposed for encrypted traffic categorization. The contemporary techniques such as dilated convolution and residual connection are adopted as the basic building block. The raw traffic files are pre-processed to generate 1-dimensional flow byte sequences and are feed into our specially-devised network. The proposed approach outperforms other existing methods greatly on a public traffic dataset.
ER -