Segurança em redes sem fio: estudo sobre o desenvolvimento de conjuntos de dados para comparação de IDS

• Autor(a) principal: Vilela, Douglas Willer Ferrari Luz [UNESP]
• Data de Publicação: 2014
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://hdl.handle.net/11449/124423; http://www.athena.biblioteca.unesp.br/exlibris/bd/cathedra/17-06-2015/000836349.pdf
• topic: Redes de computadores - Medidas de segurança; Sistemas de comunicação sem fio; Sistemas de segurança; Algoritmos; Computer networks -- Security measures

The fast growth of wireless network technology has been very significant lately, its occurs in diverse sectors of society. The standard IEEE 802.11 stands out in this scenario. However, the protection mechanisms employed by this standard wireless network has not shown effectiveness in combating denial of service attacks. The intrusion detection systems are seen as an effective way to minimize these threats. We proposed in this research to build three data sets, which represent traffic wireless network. The sets are generated auxiliary purpose in assessing intrusion detection algorithms for wireless networks. For the construction of the data sets three scenarios of wireless networks, all in real operational environments and have been implemented. In each scenario was one enabled security mechanisms: WEP protocol scenario 1, scenario 2 was used IEEE 802.11i scenario 3 the associated IEEE 802.11i amendment to the IEEE 802.11w. The choice of different sets of scenarios and divide according to the environments aims to analyze the evolution of the security mechanisms. This makes it possible to categorize each environment. After the construction of wireless network environments normal and anomalous traffic were inoculated and thus collect the data. With the collected data pre-processing each set only extracting the frames from the MAC header was conducted. The choice was defined as this has specific characteristics of wireless networks. To validate the data sets and sorting algorithms were employed pattern recognition. The algorithms were used in the validation MLP, RBF and Bayes Net. The results obtained from the evaluation of the generated data sets demonstrate that the proposed approach is quite promising