Artificial Intelligence and Machine Learning in 5G Network Security: Opportunities, advantages, and future research trends

TL;DR

This paper reviews how AI and ML can enhance 5G network security by providing advanced threat detection and automation, addressing new vulnerabilities introduced by 5G's softwareized and virtualized architecture.

Contribution

It offers a comprehensive overview of AI and ML applications in 5G security, highlighting potential research directions and key data collection points for threat detection.

Findings

AI and ML improve threat classification accuracy in 5G networks

Identification of key data points for anomaly detection in 5G architecture

Discussion of future research trends in AI-driven 5G security

Abstract

Recent technological and architectural advancements in 5G networks have proven their worth as the deployment has started over the world. Key performance elevating factor from access to core network are softwareization, cloudification and virtualization of key enabling network functions. Along with the rapid evolution comes the risks, threats and vulnerabilities in the system for those who plan to exploit it. Therefore, ensuring fool proof end-to-end (E2E) security becomes a vital concern. Artificial intelligence (AI) and machine learning (ML) can play vital role in design, modelling and automation of efficient security protocols against diverse and wide range of threats. AI and ML has already proven their effectiveness in different fields for classification, identification and automation with higher accuracy. As 5G networks' primary selling point has been higher data rates and speed, it will be difficult to tackle wide range of threats from different points using typical/traditional protective measures. Therefore, AI and ML can play central role in protecting highly data-driven softwareized and virtualized network components. This article presents AI and ML driven applications for 5G network security, their implications and possible research directions. Also, an overview of key data collection points in 5G architecture for threat classification and anomaly detection are discussed.