Edge Caching Strategy Based on User Preference and Game Theory

Abstract

Caching video content to the mobile edge server is an effective solution to avoid multiple repeated requests from mobile terminal devices, reduce latency costs, and improve user QoE. In addition, mobile users in nearby areas tend to request the same video resource task, so reasonable deployment of video content to edge servers can effectively reduce the response time of user requests. This fact prompted us to design a caching strategy based on user preferences. The system model considered in this article contains multiple mobile users, multiple servers, and remote central servers. Based on the recommendation system predicting the user’s preference for specific video resources, the recommended value ranking of the video resources in the future time period is obtained, and then based on the game theory method, in each edge cache server, each video resource is calculated for the local area and neighboring areas. The cache value of the area, based on the cache value of the video resource to be cached, minimizes the delay for users to obtain the video resource, and obtains the optimal video resource cache distribution strategy. The simulation experiment results show that compared with other caching strategies, the proposed caching strategy is better than other caching strategies in terms of response time and cache task hit rate.