A Mobile Application-Classifying Method Based on a Graph Attention Network from Encrypted Network Traffic
Classifying mobile applications from encrypted network traffic is a common and basic requirement in network security and network management. Existing works classify mobile applications from flows, based on which application fingerprints and classifiers are created. However, mobile applications often...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-05-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/12/10/2313 |
| _version_ | 1797600372370440192 |
|---|---|
| author | Guoliang Xu Ming Xu Yunzhi Chen Jiaqi Zhao |
| author_facet | Guoliang Xu Ming Xu Yunzhi Chen Jiaqi Zhao |
| author_sort | Guoliang Xu |
| collection | DOAJ |
| description | Classifying mobile applications from encrypted network traffic is a common and basic requirement in network security and network management. Existing works classify mobile applications from flows, based on which application fingerprints and classifiers are created. However, mobile applications often generate concurrent flows with varying degrees of ties, such as low discriminative flows across applications and application-specific flows. So flow-based methods suffer from low accuracy. In this paper, a novel mobile application-classifying method is proposed, capturing relationships between flows and paying attention to their importance. To capture the inter-flow relationships, the proposed method slices raw mobile traffic into traffic chunks to represent flows as nodes, embeds statistical features into nodes, and adds edges according to cross-correlations between the nodes. To pay different attention to the various flows, the proposed method builds a deep learning model based on graph attention networks, implicitly assigning importance values to flows via graph attention layers. Compared to recently developed techniques on a large dataset with 101 popular apps using the Android platform, the proposed method improved by 4–20% for accuracy, precision, recall, and F1 score, and spent much less time training. |
| first_indexed | 2024-03-11T03:47:10Z |
| format | Article |
| id | doaj.art-ddadcd778ced4d2aa37f157289e201c3 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-11T03:47:10Z |
| publishDate | 2023-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-ddadcd778ced4d2aa37f157289e201c32023-11-18T01:10:39ZengMDPI AGElectronics2079-92922023-05-011210231310.3390/electronics12102313A Mobile Application-Classifying Method Based on a Graph Attention Network from Encrypted Network TrafficGuoliang Xu0Ming Xu1Yunzhi Chen2Jiaqi Zhao3School of Cyberspace, Hangzhou Dianzi University, Hangzhou 310018, ChinaSchool of Cyberspace, Hangzhou Dianzi University, Hangzhou 310018, ChinaSchool of lnformation Engineering, Hangzhou Vocational & Technical College, Hangzhou 310018, ChinaSchool of Cyberspace, Hangzhou Dianzi University, Hangzhou 310018, ChinaClassifying mobile applications from encrypted network traffic is a common and basic requirement in network security and network management. Existing works classify mobile applications from flows, based on which application fingerprints and classifiers are created. However, mobile applications often generate concurrent flows with varying degrees of ties, such as low discriminative flows across applications and application-specific flows. So flow-based methods suffer from low accuracy. In this paper, a novel mobile application-classifying method is proposed, capturing relationships between flows and paying attention to their importance. To capture the inter-flow relationships, the proposed method slices raw mobile traffic into traffic chunks to represent flows as nodes, embeds statistical features into nodes, and adds edges according to cross-correlations between the nodes. To pay different attention to the various flows, the proposed method builds a deep learning model based on graph attention networks, implicitly assigning importance values to flows via graph attention layers. Compared to recently developed techniques on a large dataset with 101 popular apps using the Android platform, the proposed method improved by 4–20% for accuracy, precision, recall, and F1 score, and spent much less time training.https://www.mdpi.com/2079-9292/12/10/2313traffic classificationmobile application identificationencrypted trafficgraph attention networks |
| spellingShingle | Guoliang Xu Ming Xu Yunzhi Chen Jiaqi Zhao A Mobile Application-Classifying Method Based on a Graph Attention Network from Encrypted Network Traffic Electronics traffic classification mobile application identification encrypted traffic graph attention networks |
| title | A Mobile Application-Classifying Method Based on a Graph Attention Network from Encrypted Network Traffic |
| title_full | A Mobile Application-Classifying Method Based on a Graph Attention Network from Encrypted Network Traffic |
| title_fullStr | A Mobile Application-Classifying Method Based on a Graph Attention Network from Encrypted Network Traffic |
| title_full_unstemmed | A Mobile Application-Classifying Method Based on a Graph Attention Network from Encrypted Network Traffic |
| title_short | A Mobile Application-Classifying Method Based on a Graph Attention Network from Encrypted Network Traffic |
| title_sort | mobile application classifying method based on a graph attention network from encrypted network traffic |
| topic | traffic classification mobile application identification encrypted traffic graph attention networks |
| url | https://www.mdpi.com/2079-9292/12/10/2313 |
| work_keys_str_mv | AT guoliangxu amobileapplicationclassifyingmethodbasedonagraphattentionnetworkfromencryptednetworktraffic AT mingxu amobileapplicationclassifyingmethodbasedonagraphattentionnetworkfromencryptednetworktraffic AT yunzhichen amobileapplicationclassifyingmethodbasedonagraphattentionnetworkfromencryptednetworktraffic AT jiaqizhao amobileapplicationclassifyingmethodbasedonagraphattentionnetworkfromencryptednetworktraffic AT guoliangxu mobileapplicationclassifyingmethodbasedonagraphattentionnetworkfromencryptednetworktraffic AT mingxu mobileapplicationclassifyingmethodbasedonagraphattentionnetworkfromencryptednetworktraffic AT yunzhichen mobileapplicationclassifyingmethodbasedonagraphattentionnetworkfromencryptednetworktraffic AT jiaqizhao mobileapplicationclassifyingmethodbasedonagraphattentionnetworkfromencryptednetworktraffic |