State of BGP Security Against Network Anomalies Through Ensemble of Classifiers Approach

Abstract

The Internet infrastructure relies on the Border Gateway Protocol (BGP) to provide essential routing information where abnormal routing behavior interferes with global connectivity and stability. Considering the importance of the BGP network stability, a variety of methods have already been developed to identify network anomalies. Nonetheless, the biggest obstacles to detecting BGP attacks are the dynamic nature and complex structure of the network. Hence, it is difficult to classify them using a single classifier because of the dynamic nature of the network. We, therefore, suggested an approach based on an ensemble classifier to identify these anomalies. The proposed ensemble classifier uses Extreme Learning Machine (ELM), Naïve Byes (NB), and k-nearest neighbor (KNN) classifier. RIPE and BCNET, datasets are used to evaluate and compare the proposed technique. The findings of the analysis indicate that on both datasets, the proposed classifier provides better classification accuracy.