Dissertation > Excellent graduate degree dissertation topics show

Based on multi- sensor data fusion attack detection and identification of intent

Author: LinJiaRun
Tutor: YinJianPing
School: National University of Defense Science and Technology
Course: Computer Science and Technology
Keywords: Multi-sensor data fusion Extended D-S evidence theory Combo attacks Attack Detection Attack intention recognition
CLC: TP393.08
Type: Master's thesis
Year: 2010
Downloads: 104
Quote: 0
Read: Download Dissertation

Abstract

Different sensors be used to protect network security, but the each sensor be run separately has the function limitations, will produce large amounts duplicate security data and events, false alarm rate and false negatives rate is higher. Therefore, the multi sensor data fusion technology being increasingly widely used in network security fields. The current multi-sensor data fusion technology uses the data source for is more a single, in the pairs of each sensor independently generated decision results carry out fusion O'clock less consideration different sensors the advantages and disadvantages, can not fully utilize multi-sensor data's redundancy Xing and complementarity Xing. This article will multi-sensor data fusion technology applied to the network attack detection with the intention recognition medium, and its conducted in-depth research, and finally achieved its prototype system. This paper work of the main contribution to the and innovations points to summarize the as follows: 1, put forward a one kinds based on time and space network security data fusion technology classification and method of. This method to consider data fusion technology in the field of network security special the application characteristics of, from the time and space the angle of, will be used for of network security multi-sensor data fusion technology is divided into time-based data fusion techniques and space-based data fusion technology two categories of . Comparative analysis of the these two types of data fusion technology features as well as advantages and disadvantages of, discusses the used for network secure multi-sensor data fusion technology the development trend of. 2, put forward a one kinds based on multi sensor data fusion's simple Attack Detection Method. In this method of each sensor associated with logically related alarm event, generate ultra-alarm event; the fusion center will of each sensor independently generated ultra-alarm events as evidence, the use of the extended DS evidence theory carry on fusion, get a more accurate detection results; Last combined with attack relies vulnerabilities and service information, and seek of the corresponding attacks probability of success. Experimental results show that, that the method can effectively reduce the alarm events the number of, and raise awareness of a simple attack detection's accuracy rate. 3, put forward a one kinds based on extended for DS evidence theory a combination of attack detection and intent of the recognition method. The method first define the same combo attacks middle different simple attack inter-attack ordinal the concept of, proposed based on attack ordinal's attack intent of scenes representation method, and to pre-defined typical combinatorial attack intent of scenes. Finally adopt the extended DS evidence theory fusion of each sensor generates a having different credibility of the of ultra-alarm events, thus completing the combo attacks the detection and its intentions recognition. Experimental results show that, that the method can effectively detect combo attacks, and identify the attacker's attack intentions. 4, the design and implemented based on multi-sensor data fusion of the attack detection and intention recognition prototype system. The system uses modular design, has a good scalability; using a distributed's system physical structure, the in the network distributed deployment different security sensor; using hierarchical system logic structure, the will different sensors independently generated ultra-alarm events Upload to the upper layer fusion center carry out fusion; adopt Eclipse RCP (Rich Client Platform, fat client) to build systems framework and the Eclipse Zest technology showcase network topology, has a good visualizations effect of and Operability.

Related Dissertations

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