A Novel Network Attack Detection Platform Targeting the Amf Component in the 5g Network Infrastructure

Abstract In the trend of the Internet of Things, 5G technology is one of the important platforms connecting mobile devices to the Internet network. Along with the popularity of 5G network deployment in many countries, the risk of destructive attacks on this infrastructure is increasing. In this paper, a platform is proposed to support network attack detection targeting one of the important components in the 5G core, the Access and Mobility Management Function block. The proposed platform utilises a combination of local machine learning models deployed at each network node to perform real-time attack detection, along with a federated learning model that enables nodes to share internal knowledge without exchanging raw data, thereby enhancing the system's ability to detect attacks. In addition, to ensure privacy in 5G user data, the model also applies modern security solutions. The test results demonstrated that the platform's accuracy and processing speed have the potential to meet the security deployment requirements for 5G networks.