Research of Machine Learning Algorithms for the Development of Intrusion Detection Systems in 5G Mobile Networks and...

Area of Science:

• Cybersecurity
• Telecommunications Engineering
• Machine Learning

Background:

• The global rollout of fifth-generation (5G) mobile networks necessitates robust security measures against evolving cyber threats.
• Integration of new technologies into critical infrastructure raises concerns about network vulnerabilities and potential hacking.
• Existing security architectures require enhancement to address the unique challenges posed by 5G.

Purpose of the Study:

• To propose and evaluate the implementation of a machine learning-based Intrusion Detection System (IDS) within the 5G core network architecture.
• To compare the effectiveness of various machine learning (ML) and deep learning (DL) algorithms for intrusion detection in 5G environments.
• To identify the optimal ML/DL model for detecting network attacks with high accuracy.

Main Methods:

• Overview of prominent intrusion detection datasets, specifically CICIDS2017 and CSE-CIC-IDS-2018.
• Development and testing of multiple machine learning and deep learning models for network intrusion detection.
• Comparative analysis of model performance using key metrics to determine accuracy and effectiveness.

Main Results:

• Gradient Boost demonstrated superior performance among the evaluated ML and DL models.
• The Gradient Boost model achieved 99.3% accuracy on secure data and 96.4% accuracy on detecting attacks in the test set.
• The study identified specific ML algorithms as highly effective for securing 5G networks.

Conclusions:

• Machine learning-based Intrusion Detection Systems are a viable and effective component of 5G security architecture.
• Gradient Boost is a highly accurate algorithm for detecting intrusions in 5G network data.
• The proposed approach offers a significant advancement in securing next-generation mobile networks against cyber threats.