Self Adaptive Intrusion Detection Technique Using Data Mining Concept in an Ad-hoc Network

Abstract The Intrusion Detection System mainly adopted from the traditional Wired Network or Distributed Network cannot provide a satisfying solution to the increasing security threats in an Ad-Hoc Network. Adhoc-Network due to its typical infrastructure less technical background, implementing an effective security solution for it is extremely challenging. Moreover due to its vulnerable nature attackers consistently try new attack mechanism which generally goes undetected for the system that uses pattern matching to detect traces os intrusion behavior in the incoming data. The paper aims at providing IDS based on self learning technique where the system when comes across an unknown data pattern classifies it as an attack or non attack after comparing and considering its variation from an attack free scenario.