Multi-Agent-Based Anomaly Intrusion Detection

  • Authors:
  • J. Arokia Renjit;K. L. Shunmuganathan

  • Affiliations:
  • CSE Department, Jeppiaar Engineering College, Tamil Nadu, India;CSE Department, RMK Engineering College, Tamil Nadu, India

  • Venue:
  • Information Security Journal: A Global Perspective
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

Cyber security has emerged as an established discipline for computer systems and infrastructures with a focus on protecting information stored on those systems from adversaries who want to obtain, damage, corrupt, modify, destroy, or prohibit access to it. Several information security techniques are available to protect information systems against unauthorized use, duplication, modification, destruction, and virus attacks. An Intrusion Detection System IDS is a program that analyzes what happens or has happened during an execution and finds indications that the computer has been misused. In this paper, we have proposed an effective IDS in which a local agent present in every node collects data from its own system and classifies anomaly behaviors using SVM classifier. The local agent uses the mobile agent to gather information from the neighboring node to check its integrity before it allows the system to send data to its neighboring node. The local agent is also capable of removing the local system from network if the system is found to be under attack, thereby providing a global secure environment. Our system identifies successful attacks from the anomaly behaviors.