As the core technology of the next generation mobile communication system, the development of 5G key technologies needs to be able to efficiently and effectively support massive data services. Aiming at the impact of massive data traffic on mobile communication networks in 5G communication systems, this paper proposes a 5G-oriented hierarchical distributed cloud service mobile communication system architecture. The model consists of a cloud access layer, a distributed micro-cloud system, and a core cloud data center. The distributed micro cloud system consists of multiple micro clouds that are deployed to the edge of the network. The service content in the core cloud data center can be deployed and cached to the local micro cloud server in advance to reduce repeated redundant transmission of user requested content in the network. Aiming at the problem of how to determine the migration object when dynamically optimizing the resource structure, a heuristic function-based dynamic optimization algorithm for cloud resources is proposed. The experimental results show that the dynamic expansion algorithm of cloud resources based on dynamic programming ideas can better improve the performance of virtual resources, and the dynamic optimization algorithm of cloud resources based on heuristic functions can effectively and quickly optimize the resource structure, thereby improving the operating efficiency of user virtual machine groups. An efficient resource allocation scheme based on cooperative Q (Quality) learning is proposed. The environmental knowledge obtained by the base station learning and exchanging information is used for distributed resource block allocation. This resource allocation scheme can obtain the optimal resource allocation strategy in a short learning time, and can terminate the learning process at any time according to the delay requirements of different services. Compared with traditional resource allocation schemes, it can effectively improve system throughput.

作为下一代移动通信系统的核心技术，5G关键技术的开发需要能够有效地支持大规模数据服务。针对5G通信系统中海量数据流量对移动通信网络的影响，提出了一种面向5G的分层分布式云服务移动通信系统架构。该模型由云访问层，分布式微云系统和核心云数据中心组成。分布式微云系统由部署到网络边缘的多个微云组成。核心云数据中心中的服务内容可以预先部署并缓存到本地微云服务器，以减少网络中用户请求内容的重复冗余传输。针对动态优化资源结构时如何确定迁移对象的问题，提出了一种基于启发式功能的云资源动态优化算法。实验结果表明，基于动态规划思想的云资源动态扩展算法可以更好地提高虚拟资源的性能，基于启发式功能的云资源动态优化算法可以有效，快速地优化资源结构，从而改善了资源结构。用户虚拟机组的运行效率。提出了一种基于协作Q（Quality）学习的有效资源分配方案。通过基站学习和交换信息而获得的环境知识被用于分布式资源块分配。这种资源分配方案可以在较短的学习时间内获得最佳的资源分配策略，并可以根据不同业务的时延要求随时终止学习过程。与传统的资源分配方案相比，它可以有效地提高系统吞吐量。