AI-Powered Intrusion Detection and Prevention Systems in 5G Networks
Nimeshkumar Patel
Abstract
The advent of 5G technology revolutionizes connectivity, offering high-speed data transmission, ultra-reliable low-latency communication, and enhanced capacity for a multitude of devices. However, these advancements introduce complex cybersecurity challenges, as the expanded attack surface makes networks more vulnerable to intrusions and malicious activities. Traditional intrusion detection and prevention systems (IDPS) often fall short in effectively identifying and mitigating sophisticated threats due to their reliance on static algorithms and limited adaptability. To address these limitations, this research proposes an AI-powered IDPS leveraging advanced machine learning models, including Support Vector Machines (SVM), CatBoost, LightGBM, and Temporal Convolutional Networks (TCN). The proposed model focuses on real-time data processing and anomaly detection, utilizing a comprehensive dataset featuring critical network traffic attributes such as packet size, protocol type, source and destination IP addresses, duration, and threat level. The integration of these methodologies facilitates improved detection accuracy and adaptability to evolving threats. Results indicate that the TCN model outperforms other methodologies, achieving a superior accuracy of 98.61%. The model's high accuracy highlights its effectiveness in identifying both benign and malicious traffic within 5G networks, demonstrating its potential as a robust solution for modern cybersecurity challenges. By advancing the capabilities of intrusion detection systems in 5G, this research contributes significantly to enhancing network security and establishing a safer digital landscape.