基于BP神经网络的边缘缓存内容热度预测

Prediction of edge caching content heat based on BP neural network

近年来,随着5G技术和边缘计算技术的发展,为了降低回传网和核心网流量,提升用户体验,一种基于5G小基站提供边缘缓存服务的技术逐渐成为业界研究的课题。提出了基于BP神经网络的5G小基站边缘缓存内容热度预测算法,通过用户当前对互联网内容访问的频率,预测该内容未来被重复访问的可能性,用于提升缓存命中率,达到使用有限的存储资源节省最大的网络带宽的目的,提升移动边缘缓存效率。

In recent years,5G technology and edge computing technology growing rapidly.In order to reduce the traffic of backhaul and core network and improve user experience,a technology based on 5G small base station to provide edge caching service has gradually become a research topic in the industry.A prediction algorithm based on BP neural network was proposed to improve the"hot"prediction result of edge cache content from 5G small base station.This algorithm gave the possibility of future repeated access of a content through the current frequency of users’access to this content.It could be used to improve the cache hit rate,save the maximum network bandwidth with limited storage resources,and improve the efficiency of mobile edge cache.