An Artificial Immune Networking Using Intelligent Agents

The objective of the proposed work is to formally verify and implement an Artificial Immune Networking model. The work focuses on the protection against computer viruses which disrupt the normal usage of the network. The vulnerability of the network due to the malfunctions is detected in terms of faulty or malicious nodes. An Artificial Immune System or Bio-Inspired approach is adopted by making the network to automatically detect and tolerate the previously unseen normal behavior. In this system approach the dynamic description of the usual behavior of network is formally specified and clustering of decision making methods to minimize the time of detection. The theoretical analysis of such Immune Networking is also performed and the necessary network components are proposed to counter attack the intrusions. The Artificial Immune Networking and its timely response are modeled using Colored Petri Nets (CPN) at the illegal intrusions. The detection and recovery from the intrusion may be achieved by employing intelligent agents in a distributed environment. The performance of the system is determined by the correct activation of the lympho component at the earliest within the scope of intelligence given to individual agents.