A Rough Set Based Approach to Classify Node Behavior in Mobile Ad Hoc Networks

Mobile Ad Hoc Network are used in places where providing a network infrastructure is difficult. In Ad Hoc Network the mobile nodes are not controlled by any other controlling entity, they have unrestricted mobility and form the dynamic topology. This dynamically changing network topology of MANETs makes it vulnerable to many security related issues. There are some situations when one or more nodes in the network become selfish or malicious and tend to annihilate the capacity of the network. This research investigate the classification of good and bad nodes in the network by using the concept of rough set theory , that can be employed to generate simple rules and to remove irrelevant attributes for discerning the good nodes from bad nodes. Our experiment results reveals that the rough set based approach increases the network capacity and throughput of the network up to 98.9%