The main problem of deploying service function chains in virtualized 5G networks is dealt with in wireless 5G communications and effective ubiquitous learning models. The aim is to ensure differentiated network performance for a wide range of services while maximizing the collaborative revenue of infrastructure operators and wireless virtual operators. To achieve this, a utility-based service function chain deployment strategy is introduced, tailored to the specific characteristics of Ubiquitous Learning based 5G-RAN-Architecture. Consideration is given to the virtual operator's maximum tolerable end-to-end latency, minimum service rate requirements, and the infrastructure operator's constraints on computing and link resources. It also looks at how different deployment scenarios for service function chains affect network performance and creates a utility model using a business framework. The ultimate objective is to optimize the collective revenue of infrastructure operators and virtual operators. The approach leverages genetic algorithms and Matlab's Linprog function for iterative problem-solving. The graph clearly shows that the SFC deployment algorithm in this study uses less infrastructure resources for Front Haul links. This implies that the algorithm successfully reduces the load on Front Haul lines, which in turn lowers the cost of SFC deployment and makes it easier to deploy more SFCs across the infrastructure. This work contributes to the evolution of 5G wireless communications and its seamless integration with ubiquitous learning models.

中文翻译：

---

通过基于效用的服务功能链部署实现 5G 通信网络效用最大化的无处不在的学习模型

在虚拟化 5G 网络中部署服务功能链的主要问题是通过无线 5G 通信和有效的普适学习模型来解决。其目的是确保各种服务的差异化网络性能，同时最大化基础设施运营商和无线虚拟运营商的协作收入。为了实现这一目标，引入了一种基于实用程序的服务功能链部署策略，该策略是针对基于泛在学习的 5G-RAN 架构的具体特征而定制的。考虑虚拟运营商最大可容忍的端到端延迟、最低服务速率要求以及基础设施运营商对计算和链路资源的约束。它还研究了服务功能链的不同部署场景如何影响网络性能，并使用业务框架创建实用模型。最终目标是优化基础设施运营商和虚拟运营商的集体收入。该方法利用遗传算法和 Matlab 的 Linprog 函数来迭代解决问题。该图清楚地表明本研究中的 SFC 部署算法对前传链路使用较少的基础设施资源。这意味着该算法成功减少了前程线路的负载，从而降低了 SFC 部署成本，并使在基础设施中部署更多 SFC 变得更加容易。这项工作有助于 5G 无线通信的发展及其与无处不在的学习模型的无缝集成。