Edge Caching Strategy Based on User's Long and Short Term Interests
Mingyu Si, Qiang Liu, Zhiyong Zhou, Wenbo Yang · 2023
In recent years, the exponential rise of mobile traffic has imposed a significant load on base stations and mobile core networks. Consumers' low-latency caching demands are difficult to satisfy using existing caching networks. Edge caching technology, which deploys caching servers in closer proximity to users, offers an effective solution and reduces content delivery time while avoiding network congestion. The capacity of cache servers at the network edge is restricted. Therefore, how to utilize the limited cache area for customers is a crucial concern of edge caching. To address this problem, a caching strategy that considers the long and short-term interests of users is proposed. First, users' interests are predicted based on their historical request information from the perspective of long and short-term interests, and a recommendation list is generated for each user. Then, the content is stored on one or more caching servers within the region, depending on the popularity of the content in the area. Decisions regarding cache replacement are made based on the revenue generated by the content. Results from simulations demonstrate the effectiveness of the suggested edge caching strategy in reducing the average delivery latency of the content as compared to current strategies.