Loading...

Resisting Flow Content Characterization in Network Traffic

Momeni Boroojeni, Farzad | 2011

451 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 43233 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Kharrazi, Mehdi
  7. Abstract:
  8. One way to control the type of data being transferred in the network is to check the header of the packet, but in networks that have high-speed data transmission or data can move through different routes, the probability of finding the first packet which contains the header reduces tremendously, and the usage of the methods mentioned below becomes clear. So even if the packet header isn’t received correctly, or the declarative features of these flows are changed, without having exact knowledge of the data transferred, you can use a limited number of packets to obtain information such as the type and protocol being used. This is possible with the help of statistical data analysis.Traffic analysis has benefits and can solve some problems, but other than violating privacy policies, type analysis of the data exchanged in the network, allows attackers to gain important information about the network and its users, through running different algorithms. It seems like a mechanism is required to resist traffic analysis and type detection algorithms, and can communicate without any information leakage.This thesis introduces an algorithm to encode and change data taype statistical features to contribute resisting data type recognition in networks, by studying different traffic analysis methods and their strengths and weaknesses. By recoding the data exchanging between source and destination, the analyser will detect the traffic in the type sender wants. So that even in the case of conversation interception, analyser detects data in another type and no information is leaked. The results shows that the proposed algorithm has made resistant in front of traffic analysis as the traffic kinds analyser detected are what the sender wants
  9. Keywords:
  10. Network ; Traffic Analysis ; Network Security ; Traffic Analysis Resisting

 Digital Object List

 Bookmark

...see more