Multi-agents Artificial Immune System (MAAIS) Inspired by Danger Theory for Anomaly Detection

Inspired by the potential interesting ideas of the danger theory (DT), the research into artificial immune system (AIS) has been developing faster then ever. The basic role of DT embedded in AIS is to provide what T-cells should respond through the professional antigen-presenting cells when there are some cells undergoing injury, or stress or ´bad cell death´. However, in the context of AlS-based network security systems, how to minimize damage by mitigating virus infection is a key step to solve real-world problems. In this paper we present the blueprint of a DT inspired multi-agents AIS for anomaly detection, it is called MAAIS, which can dynamically adapt to various types of immune responses by configuring parameters during the course of evaluating the degree of danger signals received. Additionally, a suitable mechanism of communication between agents is employed to stipulate the better performance of the novel system.