Dissertation > Excellent graduate degree dissertation topics show

Intrusion Detection System Research and Implementation

Author: MengZhaoRan
Tutor: LiDaXing
School: Shandong University
Course: System Analysis and Integration
Keywords: Network Intrusion Detection System IDS Data Mining Rulebase
CLC: TP393.08
Type: Master's thesis
Year: 2008
Downloads: 257
Quote: 2
Read: Download Dissertation

Abstract

With the development of network technology, people's life and work are also increasingly dependent on computer networks. And because the computer network itself defects and open nature, its security has been sorely tested. Intrusion detection is an important field of computer security technology, but also the current theory of computer security research hotspot. Intrusion detection system capable of computer networks and computer systems critical node information collection and analysis, testing whether there were violations of security policy events or signs of attack, and notifies the system security administrator. Traditional intrusion detection system using pattern matching method to achieve detection, adaptability behaved in less than ideal. Intrusion detection systems most often just the collected network data and attack pattern database already compared, and found that the behavior of the security policy violation. This pattern matching intrusion detection for the known high efficiency, but for some of the latest known attack can not be accurately detected. From the use of data mining techniques to extract a large number of network connection data to judge the comparative favor the invasion of rules, you can update the database continuously updated learning intrusion rules, intrusion detection is now a hot research problem of great theoretical significance and practical value. This design, data mining techniques in network intrusion detection system application carried out further research, we propose an improved scalable intrusion detection systems, and gives the system model and processes such as architecture. It increases the data preprocessing module, the application of data mining technology to automatically extract the intrusion rule set, so that the rule base can be updated automatically. Pass the test proved that our model can meet the current network security requirements for intrusion detection system for a relatively small input change attack sensitivity, the system itself has better adaptability. In this thesis work: understand the current intrusion detection in the context of development and the status quo, based on the analysis of the traditional intrusion detection systems strengths and weaknesses. Introduced the intrusion detection system definition and a variety of detection technology, which focuses on data mining and data mining based intrusion detection model. An improved model of scalable IDS, increased data preprocessing module (the original data standardization) and data mining module (automatic extraction of intrusion rules), and designed the intrusion rule base (can be updated automatically). Improved model of the structure consists of three subsystems, divided into data acquisition, data analysis and response systems, and gives the running processes. The IDS model for improving the performance test, various attacks in the given case the test system, anti-aggressive, and the experimental data for analysis and description. By analyzing the test data show that the improved system can have better anti-attack capability, with better scalability. Concludes the main work of this paper and the problems pointed out the future direction of further research. During the reading of the main research work: July 2006 -07 August, Jinan City, Shandong Province De Computer Technology Limited Intranet module development and network security design, combined with network intrusion detection system and the company's VPN transmission and other tools, the successful development of functional modules within the network, after testing has been put into operation. During the inspection of books and a lot of information on the performance of intrusion detection systems and models have done some research.

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