A vehicular edge computing content caching solution based on content prediction and D4PG.

Traditional research on vehicular edge computing often ignores the challenges brought by the rapid movement of vehicles and the dynamic characteristics of the environment, and often ignores that different vehicles on the same path may generate the same computing tasks, resulting in a large number of repeated calculations. Therefore, this paper proposes a vehicular edge computing content caching solution using content prediction and D4PG. Considering the complexity of the vehicular edge environment, this study proposes a digital twin-assisted method to digitally simulate the vehicular edge environment to assist in the decision-making process related to traffic prediction and content caching strategy. In response to the problems of high-speed vehicle movement and dynamic environmental changes, this paper proposes an informer-based traffic prediction model, which uses the informer prediction model to predict the environment and provide information for vehicle task content caching. At the same time, considering the problem that different vehicles on the same path may generate the same computing tasks, this paper proposes a content caching model based on distributed deterministic policy gradient (D4PG), and uses the D4PG content caching model to determine the content caching strategy. Experimental results show that this scheme can effectively reduce the vehicle task processing delay. [ABSTRACT FROM AUTHOR]