A bio-inspired multi-tissues growing algorithm for IDS based on Danger Theory and data fields

The research into intrusion detection system (IDS) inspired by immunology has been moderately successful, but was still tarnished by high false positive and poor adaptation which were caused by the lack of the environmental awareness. A novel hypothesis in immunology, the danger theory (DT), is emerging to meet these challenges. DT highlights the importance of the tissue and states that antigen presentation must be coupled with its tissue conditions. In order to create a contextually aware IDS with the incorporation of the concept of tissue, a novel algorithm to grow tissue is presented which adopts the theory of data fields in physics. This algorithm has the ability to dynamically cluster antigen data and provides a useful measure to danger signal. In order to reduce the false positive and increase robustness for detection system, a notion of multi-tissues cooperation to sense dangers in IDS is also demonstrated based on the proposed algorithm.