Evaluation of Classification Algorithms for Intrusion Detection in MANETs.
Journal
Knowledge-Based Systems
Type
journal article
Date Issued
2012
Author(s)
Abstract
Mobile Ad hoc Networks (MANETs) are wireless networks without fixed infrastructure based on the cooperation of independent mobile nodes. The proliferation of these networks and their use in critical scenarios (like battlefield communications or vehicular networks) require new security mechanisms and policies to guarantee the integrity, confidentiality and availability of the data transmitted. Intrusion Detection Systems used in wired networks are inappropriate in this kind of networks since different vulnerabilities may appear due to resource constraints of the participating nodes and the nature of the communication. This article presents a comparison of the effectiveness of six different classifiers to detect malicious activities in MANETs. Results show that Genetic Programming and Support Vector Machines may help considerably in detecting malicious activities in MANETs.
Language
English
Keywords
MANET
Intrusion detection
Genetic Programming
Classification algorithms
Support Vector Machines
HSG Classification
contribution to scientific community
Refereed
Yes
Publisher
Elsevier
Volume
36
Start page
217
End page
225
Pages
9
Subject(s)
Division(s)
Eprints ID
262951
File(s)![Thumbnail Image]()
Loading...
open.access
Name
1-s2.0-S0950705112001876-main.pdf
Size
546.29 KB
Format
Adobe PDF
Checksum (MD5)
69fe4aa35cde967483d5e1b928ee98cb