SMDFbs: Specification-Based Misbehavior Detection for False Base Stations.

Area of Science:

• Computer Science
• Telecommunications Engineering
• Network Security

Background:

• Cellular technology enables advanced services like HD video streaming and autonomous driving.
• False base stations pose a significant threat to the security and sustainability of cellular networks, particularly the Radio Access Network (RAN).
• Existing detection methods may not be sufficient to counter these evolving threats.

Purpose of the Study:

• To propose and evaluate a novel system, SMDFbs, for detecting false base stations in cellular networks.
• To develop a behavior rule specification-based approach for identifying network anomalies caused by false base stations.
• To compare the performance of SMDFbs against machine learning-based detection techniques.

Main Methods:

• Derived behavior rules from normal base station operations.
• Converted these rules into a state machine for anomaly detection.
• Implemented and tested the SMDFbs system in a 5G RAN simulator.
• Compared SMDFbs with seven machine learning-based false base station detection techniques.

Main Results:

• SMDFbs achieved a detection accuracy of 98% in the 5G RAN simulator.
• The proposed system demonstrated lower computational overhead compared to the evaluated machine learning algorithms.
• Successfully identified and mitigated threats posed by false base stations.

Conclusions:

• SMDFbs offers an effective and efficient solution for detecting false base stations in cellular networks.
• The behavior rule specification and state machine approach provides a robust method for network anomaly detection.
• The system contributes to enhancing the security and reliability of 5G Radio Access Networks.