On November 12, 2021, according to the ESI database, the paper titled “Joint Computing And Caching In 5G-envisioned Internet Of Vehicles: A Deep Reinforcement Learning-based Traffic Control System”, authored by Professor Bin Hu of the laboratory research group and published in the journal IEEE Transactions on Intelligent Transportation Systems, was simultaneously selected as both an ESI 0.1% Hot Paper and an ESI 1% Highly Cited Paper. This study designed a migration-aware joint resource allocation strategy for the Internet of Vehicles (IoV). Considering the mobility of vehicles and the time-varying nature of the IoV, a vehicle resource allocation architecture integrating communication, computing and caching functions was constructed. Then, the task scheduling and resource allocation decisions were modeled as a joint optimization problem to maximize the benefits of network operators. By adding a penalty function to the optimization objective, the strategy can balance between Quality of Experience (QoE) for users and network performance. Finally, an efficient solution to the problem was proposed based on a deep reinforcement learning method with the Actor-Critic architecture. Simulation experiments based on real traffic flow data in Hangzhou show that the strategy proposed in this paper outperforms benchmark methods, and can significantly improve the service performance of the IoV.

An ESI Hot Paper refers to a paper published within the past two years whose citation count ranks among the top 0.1% globally in the corresponding discipline over the most recent two months. An ESI Highly Cited Paper refers to a paper published within the past ten years whose citation count ranks among the top 1% globally in the corresponding discipline. ESI has become one of the important evaluation tools widely used worldwide to assess the international academic level and influence of universities, academic institutions, countries, or regions.