• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Doctoral Thesis
DOI
https://doi.org/10.11606/T.76.1997.tde-25032015-111336
Document
Author
Full name
Adriano Mauro Cansian
Institute/School/College
Knowledge Area
Date of Defense
Published
São Carlos, 1997
Supervisor
Committee
Moreira, Edson dos Santos (President)
Garcia Neto, Alvaro
Geus, Paulo Licio de
Stanton, Michael Anthony
Teixeira, Cesar Augusto Camilo
Title in Portuguese
Desenvolvimento de um sistema adaptativo de detecção de intrusos em redes de computadores
Keywords in Portuguese
Redes de computadores
Redes neurais
Segurança
Abstract in Portuguese
Este trabalho apresenta o desenvolvimento de um modelo de detecção de intrusão em redes de computadores. É apresentada a construção e testes de um protótipo para um ambiente Internet. O sistema, posicionado em pontos chaves da rede, examina o fluxo de dados entre as conexões, buscando por padrões de comportamento suspeito. A identificação de tais padrões é realizada por intermédio de uma rede neural. O sistema é altamente adaptativo, uma vez que novos perfis podem ser adicionados a base de dados, e assim um re-treinamento da rede neural permite reconhecê-los. Cerca de 240 perfis de intrusão diferentes foram analisados, dos quais 117 foram selecionados como sendo relevantes para esta implementação. Acredita-se que este conjunto represente cerca de 90 por cento do número total de tipos de ataques conhecidos na Internet. Finalmente, são relatados os procedimentos de avaliação do sistema, mostrando a eficiência do método
Title in English
Not available
Keywords in English
Not available
Abstract in English
This work presents the development of a network intrusion detection model. The building and testing of a prototype in the Internet environment is described. The system, positioned at key points of the network, examines the data flow between connections, searching for patterns of suspicious behaviors. The identification of those patterns is carried out by a neural network. The system is highly adaptive, since new profiles of intrusion can be easily added to the database, and so a retraining of the neural network enables the system to consider them. About 240 different intrusion profiles were analyzed, from which 11 7 were elected as been relevant for this implementation. It is believed that this set currently represents around 90 percent of the total number of attacks profiles over the Internet. Finally the description of the system evaluation procedures is reported, and can show the efficiency of the method
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
AdrianoCansianD.pdf (9.07 Mbytes)
Publishing Date
2015-03-26
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.