High-speed threat detection in 5G SDN with particle swarm optimizer integrated GRU-driven generative adversarial...

Area of Science:

• Cybersecurity
• Network Security
• Artificial Intelligence

Background:

• 5G software-defined networks (SDN) present unique security challenges.
• Traditional security measures are insufficient for detecting sophisticated attacks in 5G SDN environments.
• Machine learning (ML) and deep learning (DL) offer promising avenues for advanced attack detection.

Purpose of the Study:

• To develop an efficient deep learning (DL) model for enhanced attack detection in 5G SDN.
• To improve the performance and responsiveness to security breaches in 5G SDN environments.
• To integrate Particle Swarm Optimization (PSO) with Generative Adversarial Networks (GANs) and Gated Recurrent Units (GRUs) for robust intrusion detection.

Main Methods:

• Developed the Particle Swarm Optimizer-Gated Recurrent Unit Layer-Generative Adversarial Network-Intrusion Detection System (PSO-GRUGAN-IDS) classifier.
• Utilized PSO to optimize GAN model weights for improved backpropagation and synthetic attack data generation via GRU.
• Trained a deep classification (IDS) model using GRU and GAN-generated data alongside real attack data.

Main Results:

• The PSO-GRUGAN-IDS model achieved a high accuracy rate of 98.4% on the InSDN dataset.
• Demonstrated superior performance compared to existing DL-based intrusion detection methods.
• Achieved a precision rate of 98%, recall rate of 98.5%, and a detection time of 2.464 seconds.

Conclusions:

• The proposed PSO-GRUGAN-IDS model offers a highly effective solution for detecting intrusions in 5G SDN.
• The integration of PSO, GRU, and GAN significantly enhances intrusion detection capabilities.
• The model's high accuracy and efficiency provide confidence in its practical application for securing 5G networks.