Behaviour Analysis Of Node In Wireless Adhoc Network Using Genetic Algorithm

The security threats for wireless ad hoc networks have been increased in two-fold way in recent years. It keeps pace with technology. One of the techniques to thwart the security threat is behavior based wireless intrusion detection. This paper deals about behavior based wireless intrusion detection and it defines a feature set that characterizes wireless node behavior. The feature set is constructed from lower layers of WLAN to profile the normal behavior of wireless node. Observing a deviation from normal or expected behavior of wireless node identifies the misbehavior. The wireless node behavior is learned by genetic algorithm and current wireless node behavior can be predicted by genetic algorithm based on the past behavior. A 3-tuple value i.e. entropy index, newness index, mismatch index is calculated for constructed feature set in a session. The 3-tuple values of a wireless node behavior in a session are compared with expected non-intrusive behavior 3-tuple value to find intrusions. The performance of wireless intrusion detection is evaluated using detection probability and false alarm probability.