Dissertation > Excellent graduate degree dissertation topics show

Study on Methods of Privacy-Preserving Multi-Stage Attack Correlation

Author: ZhangJian
Tutor: PanLi
School: Shanghai Jiaotong University
Course: Communication and Information System
Keywords: Privacy k- anonymous Alert Correlation Multi-step attack Sequential pattern mining
CLC: TP311.13
Type: Master's thesis
Year: 2011
Downloads: 25
Quote: 0
Read: Download Dissertation

Abstract


With the rapid development of network technology, global security threats growing, especially the many hazards huge multi-step attack to businesses, organizations caused serious losses, therefore, between the many organizations began to cooperate with each other to jointly resist attack . Security alarm data from various organizations often contains some data owners privacy sensitive information, such sensitive information is data owners are reluctant to publicly or share with others, which requires sensitive before these alarm data which The information protection. However, privacy protection alarm data might be a negative impact on subsequent multi-step attack associated analysis. Therefore, the study of how to balance the the alarm data privacy and usability, and how effective privacy protection technologies combined with a multi-step attack associated method is particularly important. -Depth analysis of the advantages and disadvantages of the traditional privacy protection technology and multi-step attack related methods, on this basis, the key research each realization algorithm and the two effectively combine the complete set of load Privacy multi-step attack associated with the framework and its corresponding algorithm. This paper summarizes the the ordinary data privacy protection technology, analyzes the advantages and disadvantages of several classic privacy protection technology, and focused on the k-anonymity model and its algorithm. Combined with the characteristics of alarm sensitive properties, the paper gives a k-anonymous for alarm data privacy protection method, the method is based on the ideology, the first use of the discrete attributes entropy or differential entropy of continuous attributes design reasonable alarm sensitive attribute generalization hierarchies, and then using the improved Incognito algorithm alarm data privacy protection, to better protect the privacy of alarm data security at the same time improve the efficiency of the algorithm. Secondly, the analysis summarizes several classic multi-step attack associated methods, and focuses on the multi-step attack based on sequential pattern mining association methods. For the deficiencies of the existing sequence pattern mining algorithm, the paper proposed the ESPM algorithm, the algorithm to reduce the size of the database to be excavated by increasing disposable filtering and dynamic filtering of two filter stages, modify candidate attack sequence set pruning stage to improve the efficiency of the mining algorithm. In order to reduce the false alarm rate, combined with a multi-step attack behavior characteristics, further improved the ESPM-D algorithm, the algorithm based on the destination IP alarm database is divided into several sub-alarm database ESPM algorithm alarm database of the various sub-more step attack associated. Then combined with the alarm data privacy protection algorithm, multi-step attack loaded Privacy association framework, the framework ESPM-P algorithm loaded Privacy attack scenes build in privacy protection alarm data can still dig real attack scene graph. Finally, this paper, ESPM algorithms, ESPM-D algorithm and ESPM-P algorithm for DARPA 2000 intrusion detection attacks datasets validity testing laboratories, experimental results show the effectiveness of the package of framework and algorithm proposed in this paper.

Related Dissertations

  1. Web2.0 network under the Privacy and Personal Data Protection,G350
  2. IT changes on the privacy of library users,G252
  3. Study on the Tort of Privacy Right in Network,D923
  4. Privacy dialysis \,D923
  5. Legal protection of the privacy of personal information in networks,D923
  6. Privacy in the human flesh search,D923
  7. A Study of Legislation on Monitoring Measures in China,D925.2
  8. On the protection of the privacy of the investigation stage,D925.2
  9. Civil Protection of Internet Privacy,D913
  10. The Study on the Legal System of Fiancial Pricacy,D923
  11. Mining User Traversal Sequential Patterns Based on User Traversal Interest from Web Log,TP311.13
  12. The Protection of Privacy in Network Environment,D923
  13. On the Legal Protection of Internet Privacy from Photo Scandal,D923
  14. The Study of the Privacy Rights of Celebrities,D923
  15. Research on Purpose-based Access Control in Relational Database,TP311.13
  16. Data Privacy-Preserving Schemes for Cloud Computing,TP393.08
  17. The Regulation of "the Human Flesh Search" Behavior Isbased on the Online Privacy,D923
  18. Research and Implementation on Smartcard Database Management System,TP311.52
  19. Elementary Discussion to Intuitive Sagacity of Chu Bamboo Five Elements,B222.6
  20. The Research on Implementing Real Name Registration from the Perspective of Internet Infringement,F49
  21. Some Thoughts on Legal Issues of the Real-name System in Network,F49

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer software > Program design,software engineering > Programming > Database theory and systems
© 2012 www.DissertationTopic.Net  Mobile