A Digital Twin Platform for QoS Optimization Under DoS Attacks for Next Generation Radio Networks

TL;DR

The paper tackles QoS degradation in 6G networks under DoS attacks by introducing a Digital Twin platform that enables bidirectional UE–DT interaction and AI-driven QoS management. It proposes a DT-based framework combining simulated annealing for NR numerology reconfiguration and XGBoost-based QoS forecasting, validated via an emergency-use case under UDP flood in Simu5G. Key contributions include reactive numerology optimization, real-time what-if analysis, and demonstrated QoS gains (higher packet reception, lower delay, higher throughput) as network load increases. The approach offers practical benefits for critical infrastructure and emergency communications by enabling autonomous, data-driven QoS adaptation in hostile network conditions.

Abstract

Digital Twins are being used as an enabling technology in 6G applications across various domains, valued for their data-driven insights and real-time decision-making capabilities. However, integrating Digital Twins into 6G environments presents challenges in maintaining consistent network services under adverse conditions such as including denial-of-service (DoS) attacks, while ensuring consistent Quality of Service (QoS). In this work, we present a Digital Twin Platform to facilitate bidirectional communication between User Equipment (UEs) and application-specific digital twins to enhance UE traffic under UDP flood attacks. By leveraging AI to analyze key digital twin parameters such as throughput and delay, our framework derives actionable insights that enhance QoS management in DoS attack scenarios, ultimately advancing real-world applications of digital twins in critical infrastructure domains. The performance of this Digital Twin Platform is validated through an emergency management use-case in 6G networks while the network is under attack with UDP flood attacks in terms of packet reception success rate, average packet delay, and average throughput metrics.

Figures (6)

• Figure 1: QoS-aware 6G Digital Twin approach for QoS optimization.
• Figure 2: Evaluation flow for proposed Digital Twin-based QoS optimization framework.
• Figure 3: Evaluation network overview.
• Figure 4: Comparison of packet success ratio for different scenarios.
• Figure 5: Comparison of average delays across different scenarios.