Survey of research on encrypted traffic classification based on machine learning

Encrypted traffic classification was an important component of network management and security protection. However, the complexity and variability of the current network traffic environment rendered traditional classification methods largely ineffective. Machine learning, particularly deep learning,...

Full description

Saved in:
Bibliographic Details
Main Authors: FU Yu, LIU Taotao, WANG Kun, YU Yihan
Format: Article
Language:zho
Published: Editorial Department of Journal on Communications 2025-01-01
Series:Tongxin xuebao
Subjects:
Online Access:http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2025006/
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Encrypted traffic classification was an important component of network management and security protection. However, the complexity and variability of the current network traffic environment rendered traditional classification methods largely ineffective. Machine learning, particularly deep learning, with its strong feature extraction capabilities, has been widely used in the field of encrypted traffic classification. To this end, a systematic review of the latest advancements in machine learning-driven encrypted traffic classification was provided. Firstly, the encrypted traffic classification work was roughly divided into three parts: data collection and processing, feature extraction and selection, and traffic classification and performance evaluation, which correspond to data acquisition, significant feature construction, and model application and validation in encrypted traffic classification. The content was further subdivided into seven stages: traffic collection, dataset construction, data preprocessing, feature extraction, feature selection, classification models, and performance evaluation. A comprehensive summary, synthesis, and analysis of these seven stages were then conducted. Finally, the challenges faced by current research were analyzed in detail, and the future research directions for encrypted traffic classification were prospected.
ISSN:1000-436X