The Security Evaluation of ATM Information System Based on Bayesian Regularization

Air Traffic Management (ATM) system is enticing targets of cyber-attacks since 9.11 event, and the security situation of ATM information system is closely related to the safety flight of air transportation. In this paper, an approach of security evaluation for ATM system is proposed based on artificial neural network (ANN). The proposed approach combines neural networks with Bayesian regularization to simulate ATM system in ANN. An indicator system of security situation for ATM system is established as the input of ANN, and the output is system security level, which is obtained by training the ANN with the daily recoded data of ATM system. Experimental results show that the proposed method has a certain precision and practicality.