SpiralEdge-IoV: secure and adaptive task offloading in internet of vehicles edge computing using logarithmic spiral...

Area of Science:

• Intelligent Transportation Systems
• Vehicular Edge Computing
• Network Security

Background:

• The Internet of Vehicles (IoV) is crucial for intelligent transportation systems, enabling vehicle-edge computing for demanding applications.
• Existing task offloading methods in IoV often overlook security and adaptability, failing in dynamic, adversarial environments.
• High mobility, intermittent connectivity, and malicious behavior compromise IoV reliability and security.

Purpose of the Study:

• To design a secure and adaptive task offloading framework for the Internet of Vehicles (IoV).
• To address the limitations of existing approaches by integrating defense mechanisms with optimization strategies.
• To enhance the reliability and security of vehicular edge computing in challenging operational conditions.

Main Methods:

• Developed SpiralEdge-IoV, a framework integrating a logarithmic spiral defense (LSD) mechanism for adaptive trust modeling and risk evaluation.
• Integrated LSD with a bio-inspired Addax-optimization-based decision model (LSD-AddaxNet) for security-aware multi-objective offloading.
• Evaluated the framework using realistic vehicular edge-offloading traces and the VeReMi misbehavior dataset.

Main Results:

• SpiralEdge-IoV achieved up to 18% lower average task latency and reduced energy consumption by approximately 15%.
• The framework increased task success rates in adversarial settings by over 20% compared to baseline methods.
• Demonstrated stable optimization with acceptable runtime overhead, even in dense vehicular scenarios.

Conclusions:

• SpiralEdge-IoV provides an effective solution for attack-resilient, low-latency IoV edge computing.
• The framework is suitable for safety-critical vehicular applications and future intelligent transportation systems.
• The integration of adaptive defense and bio-inspired optimization enhances IoV system robustness and performance.