New Caching Strategy within Mobile Edge Computing Architecture in Internet of Vehicles

Abstract

The Internet of Vehicles (IoV) is revolutionising transportation by enabling in- telligent, connected, and autonomous vehicles. However, the massive amounts of data generated by IoV applications place significant demands on network resources. Mobile Edge Computing (MEC) architectures, which bring data processing and storage closer to vehicles, are crucial for meeting the low-latency requirements of IoV. Caching is a key enabler for IoV-MEC systems, allowing frequently accessed data to be stored locally for faster retrieval. This thesis proposes advanced, data-driven caching strategies tailored for IoV-MEC environments. The core contribution is the development of a novel caching strategy that leverages machine learning to predict future data demands based on historical access patterns and vehicle mobility. Data is proactively cached at the optimal edge locations to minimise access latency. Extensive simulations are conducted using real-world datasets to evaluate the proposed approach against baselines. Results demonstrate significant improvements in key metrics like cache hit ratio, latency, and network load compared to existing methods. For example, the new strategy achieves a 25% higher cache hit ratio and 30% lower latency than popularity-based caching. The theoretical implications include a better understanding of the interplay between caching, machine learning, and IoV performance. Practically, the findings enable more efficient, responsive, and secure IoV systems, accelerating the adoption of autonomous vehicles.

The Internet of Vehicles (IoV) is revolutionising transportation by enabling in- telligent, connected, and autonomous vehicles. However, the massive amounts of data generated by IoV applications place significant demands on network resources. Mobile Edge Computing (MEC) architectures, which bring data processing and storage closer to vehicles, are crucial for meeting the low-latency requirements of IoV. Caching is a key enabler for IoV-MEC systems, allowing frequently accessed data to be stored locally for faster retrieval. This thesis proposes advanced, data-driven caching strategies tailored for IoV-MEC environments. The core contribution is the development of a novel caching strategy that leverages machine learning to predict future data demands based on historical access patterns and vehicle mobility. Data is proactively cached at the optimal edge locations to minimise access latency. Extensive simulations are conducted using real-world datasets to evaluate the proposed approach against baselines. Results demonstrate significant improvements in key metrics like cache hit ratio, latency, and network load compared to existing methods. For example, the new strategy achieves a 25% higher cache hit ratio and 30% lower latency than popularity-based caching. The theoretical implications include a better understanding of the interplay between caching, machine learning, and IoV performance. Practically, the findings enable more efficient, responsive, and secure IoV systems, accelerating the adoption of autonomous vehicles.