Abstract—The Border Gateway Protocol (BGP) is the back- bone of the internet, yet its inherent lack of built-in security mechanisms leaves it vulnerable to threats such as BGP hijacking. These attacks can cause widespread internet outages and enable malicious traffic interception. In this paper, we propose a novel machine learning-based framework that leverages Long Short- Term Memory (LSTM) networks for real-time analysis of BGP update messages. Our methodology involved training the model on historical BGP data obtained from RIPE RIS and Route Views, followed by testing in a simulated environment using GNS3. Experimental results demonstrate that our approach achieves 98% accuracy in detecting illegitimate route announce- ments while maintaining minimal latency. Compared to existing security mechanisms such as RPKI and BGPsec, our framework provides a more efficient and scalable solution for securing inter- domain routing. Index Terms—BGP, BGP hijacking, anomaly detection, ma- chine learning, LSTM, internet security
N Hariharan (Tue,) studied this question.