IMR: MT: xGTracker -- Mobile xG Performance Monitoring and Data Collection Platform to Enable Large-Scale Crowd-Sourced Measurement

商用5G网络正变得越来越广泛。它们的主要优势是速度快得多，支持多媒体流、VR/AR(虚拟现实/增强现实)和自动驾驶汽车等新兴应用。测量这些网络和新兴应用的性能对于了解它们的运行方式和确定需要改进的领域非常重要，尤其是对于设计下一代此类技术而言。这个合作项目汇集了密歇根大学和明尼苏达大学的研究人员，创建了一个测量5G、6G(即xG)网络性能的平台，并展示了随着时间的推移新兴应用程序的性能变化。XGTracker是一个模块化、可扩展、可配置、跨技术和以应用为中心的测量平台。首先，xGTracker有几个可配置的组件，使研究人员能够添加/更换组件。它能够选择可用的无线电频段/技术进行测量。其次，xGTracker将集成现有的真正的开源应用程序，并生成不同的工作负载来模拟其他应用程序，以收集应用程序的体验质量指标。第三，xGTracker允许基于几个参数(位置、运营商和工作负载)动态选择服务器。这有助于了解服务器位置对收集的指标的影响。第四，XGTracker将报告不同系统/设备组件的能耗，从而能够监控新兴应用的能耗。最后，xGTracker充分考虑了用户隐私，允许用户选择共享什么以及如何共享他们的数据。该项目的更广泛影响是多维的。首先，xGTracker提供了衡量和表征商业xG网络性能的工具。这将使xG客户、应用程序开发人员和xG运营商受益。XGTracker有潜力与行业合作者融合，提升未来亿万xG用户的体验质量。其次，xGTracker提供了一个整合研究和教育的机会。它将为教授的网络和移动课程贡献新的内容，并帮助设计各种课程项目。XGTracker还将被用来向学生展示5G技术，特别是来自代表不足的群体的学生，并模拟他们对STEM的兴趣。XGTracker平台的存储库位于https://github.com/xGTracker-Platform.该项目预计将在BSD风格的许可自由软件许可下开放源码，供其他研究人员贡献和添加新功能。通过xGTracker平台收集的数据将通过不同的性能图提供，以显示不同技术随时间的性能以及不同应用程序的体验质量。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Commercial 5G networks are becoming more widely available. Their key advantage is much higher speeds, enabling emerging applications such as multimedia streaming, VR/AR (virtual reality/augmented reality), and autonomous vehicles, to name a few. Measuring the performance of these networks and emerging applications becomes important to understand how they function and to identify areas of improvement especially for designing the next generation of such technologies. This collaborative project brings together investigators from the University of Michigan and University of Minnesota to create a platform for measuring the performance of 5G, 6G, (i.e., xG) networks and show the changes over time along with the performance of the emerging applications. xGTracker is modular, extensible, configurable, cross-technology, and application-centric measurement platform. First, xGTracker has several configurable components and enables researchers to add/replace components. It is capable of selecting available radio bands/technologies to conduct measurements. Second, xGTracker will integrate existing real open-source applications as well as generating different workloads to emulate others for collecting application Quality of Experience metrics. Third, xGTracker allows for dynamic server selection based on several parameters such as (location, carrier, and workload). This helps understand the impact of server location on the collected metrics. Fourth, XGTracker will report energy consumption for different system/device components which enables monitoring the energy consumption for emerging applications. Finally, xGTracker fully considers user privacy allowing users to choose what and how to share their data.The broader impact of the project has multiple dimensions. First, xGTracker provides tools that measure and characterize the performance of commercial xG networks. This will benefit xG customers, application developers, and xG carriers. XGTracker has the potential to be integrated with industrial collaborators improving the quality of experience for hundreds of millions of xG users in the future. Second, xGTracker presents an opportunity to integrate research and education. It will contribute new content to networking and mobile courses taught and help design various course projects. xGTracker will also be used to show 5G technology to students especially from underrepresented groups and simulate their interest in STEM. The repository for the xGTracker Platform is at https://github.com/xGTracker-Platform. The project is expected to be open-source under a BSD-style permissive free software license for other researchers to contribute to it and add new features. The data collected through the xGTracker platform will then be made available through different performance maps to show the performance of the different technologies over time as well as the quality of experience for different applications.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.