www.ijraset.com
Vol. 1 Issue V, December 2013 ISSN: 2321-9653
INTERNATIONAL JOURNAL FOR RESEARCH IN AP PLIED SCIENCE AN D E N G I N E E R I N G T E C H N O L O G Y (I J R A S E T)
Analysis of Artificial Neural Networks Based Intrusion Detection Systems for Mobile Ad Hoc Networks Alka Chaudhary#, V.N.Tiwari*, Anil Kumar# #
CSE, Manipal University jaipur, India
*
E&C, Manipal University jaipur, India
Abstract— In mobile ad hoc networks, Intrusion detection system is known as the second line of defense because prevention based techniques are not a good solution for ad hoc networks due to its complex characteristics. For the security point of view, many intrusion detection systems have been proposed to mobile ad hoc networks in literature. This paper analyzed the proposed artificial neural networks based intrusion detection systems and also discussed their applicability in mobile ad hoc networks. Keywords— Mobile ad hoc networks (MANETs), MANETs Security issues, Intrusion detection system (IDS), IDS components, artificial neural networks (ANNs).
I. INTRODUCTION
Mobile ad hoc networks are very flexible network in terms of communication because there is no need of predefine infrastructure for communication between the mobile nodes. Some
characteristics of MANETs such as communication via wireless links, resource constraints (such as bandwidth and battery power), cooperation between the nodes due to communication protocols and dynamic topologies make it more vulnerable to attacks [1]. Intrusion detection is a security technology that attempt to identify those who are trying to break into and misuse a system without authorization and those who have legitimate access to the system but are abusing their privileges. An Intrusion detection system dynamically monitors a system and user actions in the system to detect intrusion [2]. The basic functionality of IDS depends only on three main components such as data collection, detection and response. The data collection component is responsible for collect the data from various data sources such as system audit data, network traffic data, etc. Detection module is responsible to
analysis of collected data to detect the intrusions and if detection module is detected any suspicious activity in the network then initiates the response by the response module. There are mainly three detection techniques such as misuse based, anomaly based and specification based techniques presented in the literature [3]. Misuse-based detection systems detect the intrusions on the behalf of predefined attack signature. Second intrusion detection technique is Anomaly-based detection technique. It detects the intrusion on bases of normal behavior of the system. The third technique is specification - based intrusion detection. In this detection technique, first specified the set of constraints on a particular protocol or program and then detects the intrusions at the run time violation of these specifications. This paper emphasized on proposed neural networks based intrusion detection system in mobile ad hoc networks. Artificial neural networks are very important method to differentiate normal and malicious nodes in the MANETs. ANNs have the ability to detect intrusions with high detection rate and forecasting the unknown patterns or learning capability [7]. In the rest of paper are organized as
Page 70