Mobile Edge Cache Strategy Based on Neural Collaborative Filtering

Abstract

In order to effectively reduce the network transmission delay and improve the network transmission quality, the concept of Content Delivery Network (CDN) is brought forth to provide necessary technical support. In this paper, the edge cooperative caching (ECC) based on machine learning and greedy algorithm is put forward. To start with, the neural collaborative filtering is used to design the content popularity prediction algorithm to realize more accurate prediction of content popularity. Following that, the greedy algorithm after optimization is used to obtain the content delivery strategy of various servers in the cooperative cache domain. Finally, the ECC is adopted to achieve the optimization goal of minimal average content transmission delay. Meanwhile, the simulation experiment is carried out to verify the performance of the ECC. The experimental results suggest that the ECC can effectively improve the cache hit rate and the content cache space utilization, and shorten the average content transmission delay.

Keywords

• Collaborative filtering
• neural network
• content distribution network
• mobile edge caching
• proactive caching