SM-RC: A New Security Measurement Method for Inter-Domain Routing System

Abstract

Inter-domain routing system (IRS) is the key infrastructure of the Internet. It is of great significance to perceive and identify the security threats of IRS for the stable operation of the Internet. Security quantification is an important part of security threat perception. The existing methods are unable to construct accurate abnormal state quantitative models because of limited data. Therefore, only the similarity between the current state and the normal state of IRC can be considered, and the types of security threats cannot be distinguished. Aiming at the above problems, this paper proposes a security measurement method based on the abnormal state modeling. By analyzing the security indicators, when IRS suffering from security events and extracting the common characteristics of the abnormal state, a qualitative model of IRS's abnormal state is established. Thus, the similarity between the current state and the specific abnormal state has been taken into consideration in measuring the security of IRS. Experiments based on SMW-4 cable fault event and Malaysian route leak event show that the highest detection and classification precision of the former are both 100%, and the highest detection and classification precision of the latter is 96% and 95.8%, respectively, which proves that the measurement results can indicate the occurrence of security threats and their types.