Intelligent Analysis for Abnormal Signaling Detection on NGN Voice Call
Wenlin Cheng · 2022
Due to the maturity of Next Generation Network (NGN) technology, traditional Public Switched Telephone Network (PSTN) and mobile core networks have been replaced by IP-based networks. Nowadays, both fixed and mobile phone core networks are migrated to NGN to provide high-quality services. For VoIP calls in NGN, this research determines whether the call is interrupted by analyzing the packet content. Observation through packet monitoring shows the failures do not occur frequently. In most cases, communication is normal. Therefore, we tried to solve the data imbalance problem during the collection of obstacle cases. After a significant increase in the number of obstacle cases, we built three AI models to analyze signaling handshake packets for quickly detecting communication failures and improving the quality of VoIP.