基于神经网络的自适应无线局域网入侵检测系统研究-计算机软件与理论专业论文.docxVIP

Abstract Abstract ABS丁RACT With the rapid development of wireless local area network ，the existence of its own security problems ，has also slowly attracted attention. Intrusion detection system has become a research focus of network secl1rity as an important component of inforrnation secl1rity. Although research on cable network intrusion detection system has been relatively mature，further research on intrusion detection technology specific to wireless local area network is still required because we cannot apply existing technologies dírectly to wireless networks due to differences between wireless networks and cable networks. 丁raditional Intrusion Detection technologies are based mainly on rules and expert knowledge base and are strongJy dependent on them relatively，and thus does not adapt well to the environmen t. As a result ，this article focuses on the work of introducing the neural network technology into the wireless LAN intrusion detection system，to improve the accuracy of intrusion detection systems and real-time by improvíng the traditíonal neural network algorithm. This article is outlined as the following: Analysis the characteristics of wireless LAN security based on the full research of existing security vulnerabilities and attacks. Improve the traditional BP neural network algorithm ，Propose the BP neural network algorithm based on the minimum transition ，and then design detection model based on neural network algorithm by analyzing the characteristics of wireless LAN security holes. Finally，design distributed multi-Agent wireless LAN intrusion detection system， add detection model based on neural network algorithm to the system，analysis the results of the simulating attacks test on the model. Key Words: wireless lan ，intrusion detection neural network ，information security II 中国科学技术大学学位论文原创性声明 本人声明所呈交的学位论文，是本人在导师指导下进行研究工作所取得的 成果。除巴特别加以标注和致谢的地方外，论文中不包含任何他人已经发表现 撰写过的研究成果。与我一间工作的问志对本研究所做的贡献均已在论文中作 了明确的悦明。 作者签名