A novel MEC-enabled Blockchain-based System Architecture for Smart Vehicles Data Privacy: A Deep Reinforcement Learning Approach

Over the past few years, there has been considerable interest in the research of the Internet of Things (IoT) and Artificial Intelligence (AI). IoT is considered to be a component of the future internet and will be composed of billions of intelligent and communicating devices. The internet of the future will be made up of variously connected devices that will expand the boundaries of the world by integrating physical objects with virtual components. The combination of AI and IoT is driving the development of automatic transportation systems and the concept of intelligent cars and smart vehicles in the present century. With the increasing adoption of AI and IoT in smart vehicles, the potential security and privacy challenges are amplified. As these technologies become more integrated into vehicle systems, ensuring robust cybersecurity measures and safeguarding personal data become paramount to protect against potential threats and maintain user trust. In this paper, we propose a novel system architecture that combines Blockchain technology, Mobile Edge Computing (MEC), and Deep Reinforcement Learning (DRL) for enhancing smart vehicles data privacy. Our proposed architecture aims to address the challenges mentioned, by leveraging the decentralized and immutable nature of blockchain, the computational capabilities of MEC, and the intelligent decision-making of DRL and Neural Networks. The results of the combined experiment are compared with DQN and experimental results demonstrate a significant improvement of ۱۰.۲% in privacy level and an average of ۱۵.۱% in latency with the proposed scheme.

Keywords:

Smart vehicles , Mobile edge computing , Blockchain , Task Offloading , Deep reinforcement learning.