Loading...

Enhancing the Confidentiality of Encrypted Traffic with the Adversarial-Learning Approach

Tajalli, Hamid Reza | 2020

318 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 52928 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Jalili, Rasool
  7. Abstract:
  8. The importance of confidentiality and anonymity maintaining mechanisms are not hidden to anybody these days. With the worldwide web spreading rapidly, protecting the users' data flowing through it has become one of the most critical challenges to anonymity mechanisms. Nonetheless, machine learning algorithms have shown that they can reveal some explanatory information, even from encrypted traffic. Website fingerprinting attacks are a group of traffic analysis attacks that aim to detect the website which the monitored user has already visited. The current research takes a brief survey over website fingerprinting attacks presented in recent studies plus the defenses which took devised against them. The advent of deep neural networks paved the way for the vast improvement in classifiers' performance, and they got very high accuracy in determining visited websites. With all that said, most defensive methods are not successful enough in terms of defeating new DNN classifiers, or they add too much overhead to the traffic, which is not appropriate. The current thesis proposes a novel approach that operates adversarial machine learning concepts. By creating adversarial traffic examples from original traffic sequences, we tried to minimize the accuracy and performance of the adversary's classifier with the lowest possible overhead. The results have shown that with lower than 25% overhead, our crafted examples can degrade the targeted models' accuracy down to lower than 6% and 22% in the white box and semi-black box scenarios, respectively
  9. Keywords:
  10. Deep Learning ; Adversarial Example ; Traffic Analysis ; Website Fingerprinting ; Confidentiality ; Anonymity Network

 Digital Object List

 Bookmark

No TOC