Research On Intrusion Detection Method Based On Artificial Immune

Nowadays,network intrusion incidents occur frequently,and the network security issues receive more and more attention.Intrusion detection has quickly become a research hotspot.Because the immune recognition,immune response,immune adaptive regulation and other characteristics of the biological immune system have high similarity with the principle of intrusion detection,the artificial immune-based intrusion detection method has received extensive attention from experts and scholars.At present,the intrusion detection technology has a low detection accuracy and high false positive rate.The artificial immune theory can identify and prevent the cyber attack by imitation of the human immune system,thus ensuring network security.Applying the artificial immune theory to the intrusion detection system can improve the detection capability of intrusion detection.This paper proposes an artificial immune-based intrusion detection method through study of artificial immune theory.The main work of this paper is as follows:1)To improve the initial detector generation method,an improved initial detector generation algorithm is proposed.The clustering algorithm is used to divide the data,and the rough set algorithm is introduced.Through the calculation of the importance of the genetic attributes,the redundant attributes are identified and deleted.Finally,a rule set is generated,and the rule set is used as an initial detector to participate in intrusion detection.2)An vaccine extraction and vaccination strategy were proposed.A good individual is extracted from the memory antibody,and a gene having the gene attribute importance of more than 0.5 is extracted as a vaccine by calculating the importance of the gene attribute.The vaccine is inoculated into the immature cell antibody in a random manner,and the vaccination effect is judged by the affinity addition value to ensure that the superior gene can be inherited.3)An artificial immune-based intrusion detection model is designed.The improved initial detector generation algorithm acquires the rule set and updates it dynamically.The "self" rule set and the "non-self" rule set are respectively detected for abnormality detection and misuse detection.Immature cells are treated by genetic,mutation,and vaccine operators to ensure antibody population diversity and increase immature antibody maturation rate.By applying the KDD CUP99 dataset design simulation experiment,the experimental results show that the artificial immune-based intrusion detection method has higher detection rate and lower false positive rate than other method.