A multi-layer network defense system using artificial immune system

Abstract

Artificial immune systems are computational intelligent systems inspired by human biological immune systems. The biological immune system can detect and defend the human body against different kinds of intruders. These systems are distributed, adaptive and multilayered in nature, and the proposed artificial immune system incorporates these powerful and diverse set of characteristics in designing a network security system. Recent advances in biological immune systems showed that the innate part of the system plays a greater role than previously thought. In this paper, this recent development is the corner stone of the new multi-layer model of artificial immune systems. We describe a work in progress to construct a multi-layered defense system composed of two main layers, innate and adaptive layers. This paper describes the multi-layered system layers and components, gives a detailed justification and design considerations of the system components, and shows first results from the innate layer of the system. The preliminary results show that about 70% of the traffic can be dealt with before entering the adaptive layer of the immune system.