当前移动网络流量呈现爆发式增长态势，给移动核心网络资源利用带来巨大冲击，移动边缘缓存与分发技术作为一项有效解决途径受到广泛关注。然而现有方法难以适用于未来移动网络边缘环境复杂化，服务种类多样化，用户需求动态化等特点，对快速增长的移动内容服务质量难以保障。为此，本项目研究以用户体验（Quality of Experience, QoE）质量为驱动，深度学习为中心的移动网络边缘智能缓存与分发机制，包括（1）建立复杂网络环境下认知增强的QoE感知模型，以获取移动网络环境-用户-服务深度关联信息；（2）研究基于深度学习的动态协同边缘缓存策略，实现对移动边缘缓存资源的优化分布；（3）设计QoE驱动的智能自主分发机制，完成对移动网络多样化服务的自适配高效分发。通过上述研究，实现移动网络智能化的边缘缓存与分发，在缓解核心网络压力、优化边缘网络资源分布的同时，为大规模、多样化的高质量移动服务提供有力保障。

Recently, the mobile network traffic presents an explosive growth trend, which brings a huge impact on the utilization of mobile core network resources. As an effective solution, mobile edge caching and delivery technology has attracted wide attention. However, the existing methods are difficult to be applied in the future mobile network because of the complex edge environment , diversified service types, dynamic user demand and other characteristics, leading to be difficult to guarantee the service quality of rapid growth mobile content. For this reason, this project proposes an intelligent caching and delivery mechanism of mobile edge network, which is driven by Quality of Experience (QoE) and centered by deep learning. Including (1) establish the QoE perception model of cognitive enhancement in the complex network environment to obtain the mobile network environment-user-service depth correlation information; (2) study the dynamic collaborative edge cache strategy based on deep learning to achieve the optimal distribution of mobile edge cache resources; (3) design the intelligent autonomous delivery mechanism driven by QoE to complete the self-adaptive and efficient delivery of diversified mobile network services. Through the above research, the intelligent edge cache and delivery of mobile network can be realized, which can not only relieve the pressure of core network and optimize the resource distribution of edge network, but also provide a strong guarantee for large-scale and diversified high-quality mobile services.