Research On Airborne Network Security Protection Technology Based On Intrusion Detection

With the development of mobile Internet technology,aircraft manufacturers and airlines have designed and used a new generation of "e-enabled" aircrafts to improve the information service level.The closed airborne network interconnected with the public open network that leading to a series of airborne network security risks,such as the illegal acquisition of airborne data.The existing research on airborne network security protection technology is mostly carried out the risk assessment methods.There is no research on intrusion detection algorithm about boundary of the interaction between airborne network and public internet.This paper focuses on the intrusion detection protection technology for airborne networks.Firstly,this paper analyzes the research status of airborne network security protection technology at home and abroad,and introduces the division of airborne network domain and the composition of airborne information system,in addition,exploreing the risks and methods of attacking about the airborne network.The intrusion detection environment of the airborne network is summarized and analyzed.It is proposed to deploy the intrusion detection system at the boundary between the passenger entertainment information domain and the open domain.The experimental simulation data uses the standard commercial Ethernet dataset.Secondly,in order to solve the problem to obtain tagged data,a semi-supervised intrusion detection algorithm based on deep generation model was proposed.Variational Auto-Encoder was employed in mapping the vector of raw data from the high-dimensional space to low-dimensional,and the corresponding optimal low-dimension representation of raw can be obtained.Then,the generative model is used to improve the classification accuracy only using the tagged data.Lastly,in order to solve the problem of low detection rate about User to Root and Remote to Local and the,an semi-supervised abnormal detection algorithm was proposed.Firstly,the K-means initial clustering center was determined based on Kd-tree.Secondly,the clustered data is processed by the Tri-training method,and the tagged dataset is expanded.Finally,the hierarchical classification model is proposed by means of the binary tree model.It is used to improve the detection rate of both R2 L and U2 R attack types.