Intelligent Content Caching and User Association in Mobile Edge Computing Networks for Smart Cities

To support rapidly increasing multimedia services of smart cities, mobile edge computing (MEC) networks can significantly reduce content acquisition latency. However, due to user mobility and the possibility of re-association, it is challenging to obtain a proper content caching and user association policy. In this article, we investigate the issue of joint content caching and user association with high user mobility in MEC networks by minimizing content acquisition latency and handover latency. To address the problem, we optimize the original mixed time-scale problem in two stages: long-time scale content caching and short-time scale user association. we propose an intelligent content caching framework based on a weighted latent factor model to determine content caching policy at each time frame. Then we design a matching theory-based lazy re-association strategy at each time slot. Simulation results on real-world MEC networks demonstrate the effectiveness of the proposed framework.