HMM-AD: Anomaly Detection for 5G Control Plane based on HMM
Qian Lai Sun, Lin Tian, Jie Zeng, Miaoshun Lu · 2023
As industries increasingly adopt data-driven solutions, the deployment of fifth-generation (5G) mobile communication networks has become an urgent priority. However, recent studies have revealed that 5G network protocols are vulnerable to attacks that can lead to serious consequences such as network crashes. Furthermore, traditional security tools designed for server-class machines may not be directly applicable to the 5G control plane (CP). To address this issue, we propose a novel anomaly detection scheme for the 5G CP based on Hidden Markov Models (HMMs). Our proposed solution offers efficient and effective intrusion detection capabilities and can be deployed in 5G network servers. Additionally, we designed an anomaly-based detection algorithm and demonstrated its effectiveness through two proposed theorems. Our prototype is based on a self-developed 5G system, and the data set of signals used in our experiments is openly available on a public platform.