Loading…
Deep learning for encrypted traffic classification and unknown data detection
Despite the widespread use of encryption techniques to provide confidentiality over Internet communications, mobile device users are still susceptible to privacy and security risks. In this paper, a novel Deep Neural Network (DNN) based on a user activity detection framework is proposed to identify...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Default Article |
| Published: |
2022
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/2134/21303849.v1 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1818165738722033664 |
|---|---|
| author | Omattage Madushi H. Pathmaperuma Yogachandran Rahulamathavan Safak Dogan Ahmet Kondoz |
| author_facet | Omattage Madushi H. Pathmaperuma Yogachandran Rahulamathavan Safak Dogan Ahmet Kondoz |
| author_sort | Omattage Madushi H. Pathmaperuma (4456888) |
| collection | Figshare |
| description | Despite the widespread use of encryption techniques to provide confidentiality over Internet communications, mobile device users are still susceptible to privacy and security risks. In this paper, a novel Deep Neural Network (DNN) based on a user activity detection framework is proposed to identify fine-grained user activities performed on mobile applications (known as in-app activities) from a sniffed encrypted Internet traffic stream. One of the challenges is that there are countless applications, and it is practically impossible to collect and train a DNN model using all possible data from them. Therefore, in this work, we exploit the probability distribution of a DNN output layer to filter the data from applications that are not considered during the model training (i.e., unknown data). The proposed framework uses a time window-based approach to divide the traffic flow of activity into segments so that in-app activities can be identified just by observing only a fraction of the activity-related traffic. Our tests have shown that the DNN-based framework has demonstrated an accuracy of 90% or above in identifying previously trained in-app activities and an average accuracy of 79% in identifying previously untrained in-app activity traffic as unknown data when this framework is employed. |
| format | Default Article |
| id | rr-article-21303849 |
| institution | Loughborough University |
| publishDate | 2022 |
| record_format | Figshare |
| spelling | rr-article-213038492022-10-09T00:00:00Z Deep learning for encrypted traffic classification and unknown data detection Omattage Madushi H. Pathmaperuma (4456888) Yogachandran Rahulamathavan (2497186) Safak Dogan (1383819) Ahmet Kondoz (1384131) Ecology not elsewhere classified Distributed computing and systems software not elsewhere classified deep neural network encrypted traffic classification mobile applications network analysis wireless networks Distributed Computing Ecology <p>Despite the widespread use of encryption techniques to provide confidentiality over Internet communications, mobile device users are still susceptible to privacy and security risks. In this paper, a novel Deep Neural Network (DNN) based on a user activity detection framework is proposed to identify fine-grained user activities performed on mobile applications (known as in-app activities) from a sniffed encrypted Internet traffic stream. One of the challenges is that there are countless applications, and it is practically impossible to collect and train a DNN model using all possible data from them. Therefore, in this work, we exploit the probability distribution of a DNN output layer to filter the data from applications that are not considered during the model training (i.e., unknown data). The proposed framework uses a time window-based approach to divide the traffic flow of activity into segments so that in-app activities can be identified just by observing only a fraction of the activity-related traffic. Our tests have shown that the DNN-based framework has demonstrated an accuracy of 90% or above in identifying previously trained in-app activities and an average accuracy of 79% in identifying previously untrained in-app activity traffic as unknown data when this framework is employed. </p> 2022-10-09T00:00:00Z Text Journal contribution 2134/21303849.v1 https://figshare.com/articles/journal_contribution/Deep_learning_for_encrypted_traffic_classification_and_unknown_data_detection/21303849 CC BY 4.0 |
| spellingShingle | Ecology not elsewhere classified Distributed computing and systems software not elsewhere classified deep neural network encrypted traffic classification mobile applications network analysis wireless networks Distributed Computing Ecology Omattage Madushi H. Pathmaperuma Yogachandran Rahulamathavan Safak Dogan Ahmet Kondoz Deep learning for encrypted traffic classification and unknown data detection |
| title | Deep learning for encrypted traffic classification and unknown data detection |
| title_full | Deep learning for encrypted traffic classification and unknown data detection |
| title_fullStr | Deep learning for encrypted traffic classification and unknown data detection |
| title_full_unstemmed | Deep learning for encrypted traffic classification and unknown data detection |
| title_short | Deep learning for encrypted traffic classification and unknown data detection |
| title_sort | deep learning for encrypted traffic classification and unknown data detection |
| topic | Ecology not elsewhere classified Distributed computing and systems software not elsewhere classified deep neural network encrypted traffic classification mobile applications network analysis wireless networks Distributed Computing Ecology |
| url | https://hdl.handle.net/2134/21303849.v1 |