The high heterogeneity of 5G use cases requires the extension of the traditional per-component testing procedures provided by certification organizations, in order to devise and incorporate methodologies that cover the testing requirements from vertical applications and services. In this paper, we introduce an experimentation methodology that is defined in the context of the 5GENESIS project, which aims at enabling both the testing of network components and validation of E2E KPIs. The most important contributions of this methodology are its modularity and flexibility, as well as the open-source software that was developed for its application, which enable lightweight adoption of the methodology in any 5G testbed. We also demonstrate how the methodology can be used, by executing and analyzing different experiments in a 5G Non-Standalone (NSA) deployment at the University of Malaga. The key findings of the paper are an initial 5G performance assessment and KPI analysis and the detection of under-performance issues at the application level. Those findings highlight the need for reliable testing and validation procedures towards a fair benchmarking of generic 5G services and applications.

中文翻译：

---

5GENESIS方法

5G用例的高度异质性要求扩展认证机构提供的传统的按组件测试程序，以便设计和合并涵盖垂直应用程序和服务的测试要求的方法。在本文中，我们介绍了一种在5GENESIS项目的上下文中定义的实验方法，该方法旨在启用网络组件的测试和E2E KPI的验证。这种方法最重要的贡献是其模块化和灵活性，以及​​为其应用开发的开源软件，从而使该方法可在任何5G测试平台中轻量采用。我们还将演示如何使用该方法，通过在马拉加大学的5G非独立（NSA）部署中执行和分析不同的实验。该论文的主要发现是初步的5G性能评估和KPI分析以及在应用程序级别检测性能不足问题。这些发现强调了需要可靠的测试和验证程序，以公平地对通用5G服务和应用进行基准测试。