Heuristic Prefetching Caching Strategy to Enhance QoE in Edge Computing

摘要

The emergence of mobile devices and wireless services brings unprecedented traffic demand, which causes the bad quality of experience arises in traditional reactive networks, such as long loading time and loss of responsiveness. This paper presents the heuristic prefetching caching strategy in 5G networks, which prefetches content based on its historical frequency in order to improve the quality of experience. The cache of the base station is split into the proactive one and the reactive one. The proactive cache prefetches the popular content within the limit of capacity for a sum total maximum of frequency, while the reactive one caches others with the ordinary caching algorithm. At each period, we adjust the proportion of the proactive cache to minimize latency based on the idea of Simulated Annealing. Under the circumstances where all the content is predictable, our caching strategy improves hit ratio by 20%. And it reduces latency by 15% in the architecture of 400MB small base stations and even 52% with 200MB small base stations, which could enhance the quality of experience to a great degree.