USING MULTI-AGENT SYSTEMS FOR INTRUSION DETECTION IN COMPUTER NETWORKS: A GLANCE

Abstract

The development of Cyber attacks and information safety has become one of the important issues throughout the world. IDS (Intrusion detection systems) are one of the vital components in modern infrastructure in order to enforce various network rules and regulations. Intrusion Detection System is one of the extensively used systems which are used to diagnose malicious activities and various attacks on computer networks, but its present framework confronts a huge number of alerts and false positive alarms. Lots of work has been done to propose various MAS-based intrusion diagnostic techniques for handling the attack alerts, reducing them and for differentiating the real attacks from false positive attacks. This paper reviews various techniques used for intrusion detection in computer networks using multi-agent systems. The lack of accuracy should be improved by using various techniques like the Neural, Data mining and Threshold. Our aim will be to propose novel performance enhancement technique using multi-agent systems that will improve the lack of accuracy and false positive alarm generation problem in IDS with less processing time.