Research On Attacks And Intrusion Detection In Wireless Mesh Network

As a new kind of wireless network using multi-hop transmission mode, Wireless Mess Network(WMN) is hailed as "The Next Generation of Internet" and has a very good application prospect. However, the maturity of WMN is far less than the maturity of demand, there are still many problems need to research and solve. While bringing the advantages to WMN from the features of multi-hop and self-organization, it also appears some vulnerable when meeting the malicious attacks. So the security problem has been a research focus in the wireless Mesh network.In this paper, a research on the attacks and intrusion detection in WMN is presented. Firstly, the paper introduced WMN and some related notations and technologies of WMN, which are: 1) the concept, structure characteristics and shortages of WMN, 2) the HMMP routing protocol and two routing modes of it, 3) the principle of network intrusion detection and two intrusion detection models. Secondly, the summary and classification of the existing attack methods of WMN are conducted. And corresponding implementation schemes are designed according to the study of the principles of the typical attack methods. After that, as an example, this paper implements the black hole attack and direction misleading attack through programming. Thirdly, direct against the attacks in multi-hop transmission, an intrusion detection method based on the forwarding rate is proposed which using detection task partition method and safety state credibility mechanism to analyze and predict the security status of node. Then, this paper also implements the corresponding intrusion detection system after designing. Finally, the implemented attacks and intrusion detection system are tested respectively based on the designing of the corresponding test models. After that, the test results are analyzed and show that 1)the packet losing rate of communication between two point caused by the above two attack methods is above 95%, 2)the detection rate and non-response rate of intrusion detection system detecting the malicious attacks are 94.2% and 5.8% respectively. So, the results show that the implemented attacks and intrusion detection method achieve a better effectiveness.