TwinFedPot: Honeypot Intelligence Distillation into Digital Twin for Persistent Smart Traffic Security

Area of Science:

• Cybersecurity
• Artificial Intelligence
• Smart City Infrastructure

Background:

• Intelligent traffic systems (ITSs) in smart cities face significant cybersecurity challenges due to dynamic, adversarial cyber-physical environments.
• Securing digital twins (DTs) integrated with ITSs is crucial for reliable real-time management but remains difficult.
• Existing security solutions often struggle with the adaptive nature of threats in complex urban infrastructures.

Purpose of the Study:

• To propose TwinFedPot, an innovative digital twin-based security architecture for robust and adaptive cyber threat detection in ITSs.
• To develop a framework that enables effective threat detection without requiring prior attack data (Zero-Shot Learning).
• To enhance the security and privacy of smart traffic infrastructures against evolving cyber threats.

Main Methods:

• Implemented a novel architecture combining digital twins (DTs) with honeypot-driven data collection and Zero-Shot Learning (ZSL).
• Utilized Inverse Federated Distillation (IFD) to train the central DT server, with edge honeypots sending semantic predictions (soft logits) instead of raw data.
• Reversed the conventional federated learning knowledge flow, distilling collective intelligence from honeypots into a central teacher model.

Main Results:

• Demonstrated significant improvements in cyber threat detection accuracy and F1-score compared to conventional methods.
• Showcased the ability to learn generalized attack patterns effectively using limited data.
• Validated the framework's scalability and robustness for smart traffic infrastructures.

Conclusions:

• TwinFedPot offers a scalable and effective defense solution for securing digital twins in intelligent traffic systems.
• The proposed architecture enhances cybersecurity by enabling adaptive threat detection and preserving data privacy.
• This approach represents a significant advancement in securing smart city infrastructures against cyber-attacks.