A Content Caching Strategy Based on Demand Estimate Function in Small Cell Networks

In small cell networks, content caching is considered as an effective approach to reduce the traffic and increase the quality of experience of users. Because the knowledge about popularity distribution of content files is not available in advance, one of the main challenges in content caching is how to identify the most popular files and cache them into local limited storage on small base stations. In this paper, we propose an architecture for content caching system, a novel demand estimate function to work out the content popularity, and a cache placement strategy based on the demand estimate function. We develop a simulator implementing the proposed caching architecture and strategies to evaluate their performance. The simulation results show that the proposed strategy can achieve the stable state much faster than others. Moreover, we investigate the impact of system parameters such as cache size and the number and concentration of requests on the cache hits ratio, and the results confirm that our strategy outperforms the reference ones.