Loading...

Encrypted Traffic Analysis of a Messaging Service

Alimadadi, Zahra | 2018

639 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 50542 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Jalili, Rasool
  7. Abstract:
  8. Nowadays, with the increasing growth of internet’s speed and accessibility, and the explosion of potentials and services that smartphone applications offer, smartphones have become an inseparable part of most people's lives. At the same time, free messenger applications attract lots of people since they reduce costs, increase speed, and make communication much easier, in a way that almost all smartphone users utilize at least one of these messaging services. Moreover, the emerging use of messaging services has triggered privacy concerns and personal data treatment for users. In recent years, lots of studies focused on identifying users’ activities on the encrypted traffic, mostly with the help of machine learning techniques. Notwithstanding, the most effort has been done for a limited number of users’ activities. In this thesis, a new framework is proposed which is developed in order to investigate users’ activities in both active and passive approaches so that it can identify the most fine-grained activities in an encrypted traffic. Moreover, a scalable and automatic method for labeled dataset collection using virtual machine is introduced, which has also been used by the proposed framework. With considering the possible interaction with an international research group, Telegram was chosen as a messaging app to be analyzed in this thesis. Train and test data sets is gathered via a proposed solution separately on different virtual machines and accounts. According to the Train and test data sets is gathered via a proposed solution separately on different virtual machines and accounts. result of this research, it is not possible to identify a significant portion of user activities especially passive ones. However, it is possible to identify a few activities with respect to maintaining the privacy of message content
  9. Keywords:
  10. Traffic Analysis ; Encrypted Traffic ; Privacy ; Messaging Service

 Digital Object List