A security architecture based on immune agents for MANET

In this paper, a security architecture is designed based on the artificial immune system by using multi-agents in mobile ad hoc networks. There are two types of immunity-based agents: one is detection agent and the other is counterattack agent. Detection agents take charge of monitoring the nodes in the network, detecting the anomaly activities, and taking response to the invader, which may be viewed as T-cell lymphocyte. Counterattack agent as antibody, is built in each regular node which is regarded as B-cell. When Detection agents find invaders, they send out instructions to activate the counterattack agents in the neighbor nodes of the invaders. These Counterattack agents will surround the invaders and finally kill them. Lastly, the paper discusses how the security architecture works and protects the MANET from violating infliction by bringing forward an example.