Minimum Latency-Secure Key Transmission for Cloud-Based Internet of Vehicles Using Reinforcement Learning.

Area of Science:

• Computer Science
• Network Security
• Artificial Intelligence

Background:

• Internet of Vehicles (IoV) communication security is crucial for preventing identity-based attacks like electronic spoofing.
• Current IoV key management methods have limitations in timely group key updates, impacting security and introducing latency.
• Ensuring forward and backward security during real-time group key updates is essential as vehicles join or leave IoV clusters.

Purpose of the Study:

• To propose a low-latency Internet of Vehicles (IoV) group key distribution management technology using reinforcement learning.
• To optimize group key updates based on dynamic factors like surrounding vehicle numbers and update records.
• To reduce encryption and decryption delays in IoV communication.

Main Methods:

• A reinforcement learning-based approach for IoV group key distribution management.
• Optimization of group owner vehicle selection considering real-time environmental factors.
• Implementation of an access-driven cache attack model to minimize encryption/decryption latency.
• Simulation using Advanced Encryption Standards (AES) for verification.

Main Results:

• The proposed technology significantly reduces transmission delay for key updates in IoV networks.
• Demonstrates a reduction in calculation delay for encryption and decryption processes.
• Enhances the overall confidentiality of group keys within the IoV cluster.
• Outperforms benchmark group key management schemes in simulation.

Conclusions:

• Reinforcement learning offers an effective solution for low-latency IoV group key management.
• The proposed method enhances IoV communication security and efficiency.
• This technology provides a robust framework for real-time security in dynamic IoV environments.