Dissertation > Excellent graduate degree dissertation topics show

Research on Network Anomaly Intrusion Detector Based on Extended Dempster-Shafer Evidence Fusion Algorithm

Author: ZhaoPeng
Tutor: ZhangHuaZhong
School: Shandong University
Course: Applied Computer Technology
Keywords: D - S evidence theory Expand Merge Anomaly Detection Intrusion Detection
CLC: TP393.08
Type: Master's thesis
Year: 2008
Downloads: 138
Quote: 5
Read: Download Dissertation

Abstract


The wide application of computer network and hacker attacks frequently makes people pay more attention to network security issues. Intrusion detection technology as an important means of protecting the computer and network security, to become a research hotspot in recent years, the field of network security. With the development of the means of attack complex and the size of the network, network-based intrusion detection play an increasingly greater role. Monitoring, early warning, identification, decision-making and response network intrusion detection system by executing a series of tasks to complete the network against process, has become an important part of the network security system works. Currently, network intrusion detection is still hot and difficult intrusion detection field of study, the detection rate is not high enough, the detection range is not comprehensive enough, detection efficiency can not meet the requirements of real-time detection of large-scale high-speed network. In no guidance network intrusion detection field, DS evidence theory-based network intrusion detection technology has attracted many domestic and foreign scholars, but most remain in the application of classic DS evidence theory of network characteristics data fusion; However, the network data inevitably there is a conflict classic DS theory of evidence that there is a serious conflict fusion they can not get reasonable results, and therefore lead to the detection system, the rate of false positives and false negatives ratio were higher. In this paper, the classical Dempster-Shafer theory of evidence and Fabio extended DS evidence fusion theory, propose a evidence fusion algorithm EDS. The algorithm can be real-time evidence of a large number of serious conflict integration and be able to get a more reasonable conclusion; recognition under the framework of the two mutually exclusive objectives, the time complexity of the algorithm is only O (n), with high fusion efficiency can be applied to real-time detection of network. In view of this, EDS integration algorithm is applied to the network intrusion detection for the current lack of network anomaly detection, a real-time network intrusion detection model. This model, there is a serious conflict of network data fusion can obtain more reasonable results, thereby reducing false positives and false negatives in the model system; higher efficiency of the model checking algorithm, suitable for real-time detection of large-scale network, and having a larger detection range. The model belongs to no one guidance network anomaly detection areas, expectations of the statistical characteristics of partial variance to determine the basic probability distribution function, the salient features of rough set classification mechanism to reduce the frequency of fusion of serious conflict data to improve the accuracy of the characteristics of learning data; discrimination mechanism to reflect real-time network traffic characteristics, in order to improve the detection rate of the model system. Finally, through the UCI WBCD-dimensional data sets and the KDD Cupl999 cube experiment show that the model checking engine is based on the finite-dimensional characteristics of the data will be able to reach a higher detection rate in the lower complexity of the algorithm and lower false positive rate under the premise of , with real-time detection and good scalability, and new attacks have some immunity.

Related Dissertations

  1. Research on Joint Target Detection for Dual-Sensor Image and System Implementation,TP391.41
  2. The Study of Text Index Construction for Large-Scale Dynamic Collection,TP391.3
  3. Research of Anomaly Detection Algorithms of Hyperspectral Imagery Based on Kernel Method,TP751
  4. Intrusion detection based on the ultrasonic echo envelope in the military security patrols,E919
  5. Improving of Artificial Imune Classification and Anomaly Detection Algorithms,R392.1
  6. Research on Performance of Merge and Acquisition of Flavor and Fragrance Industry,F271
  7. Research on Intrusion Detection Technology of Wireless Sensor Networks Based on Behavior Trust,TP212.9
  8. Association rule mining based Intrusion Detection System Research and Implementation,TP393.08
  9. A Research on Expanding the Scope of VAT in China,F812.42
  10. The Research on Intrusion Detection System Based on Machine Learning,TP393.08
  11. On Ancient Roman Citizenship,K126
  12. On the Expansion in Middle-school Language Education,G633.3
  13. An Algorithm on Clustering and Anomaly Detection for Multiple Data Streams,TP311.13
  14. On the Right of Insurance Subrogation,D922.284
  15. Research on Network Anomaly Detection Based on Projection Pursuit Regression,TP393.08
  16. Research of Electronic Transformers Merging Unit,TM45
  17. Intrusion Detection in Mobile Ad Hoc Networks: A Timed Finite State Machines Approach,TN929.5
  18. An Intrusion Detection System for High-Speed Networks,TP393.08
  19. Research on the Security in Wireless Sensor Network,TN915.08
  20. HMM-based Connection between the Social Network Analysis,F49
  21. Sensitivity Analysis and Application of Orthogonal Weight Function Neural Network,TP183

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Computer network > General issues > Computer Network Security
© 2012 www.DissertationTopic.Net  Mobile