A Perception Model for Network Risk Using Danger-Theory

In order to effectively assess the risk of network risk,and to take effective prevention measures,featuring with real-time and quantitative detection,a danger-theory and Antibody-concentration based perception model for network risk is proposed.Based on dynamic self-sets,the mathematical model and dynamic equations for self-tolerance and various antibody are designed.The quantify description of antibody concentration is given.The candidate detector is generated from both random and detector genelib.Furthermore,the vaccination are introduced to enhance the active detection.Simulations are done to test the model.The experiments prove that the model can detect the intrusion attacks effectively and evaluate the risk of both host and network including each attack and the whole attacks.If the model is transformed properly,it also can be applied to the fields of virus detection and spam mail recognition.