CNS Core: Small: A Unified Approach to Internet Performance Measurement

CNS 核心：小型：互联网性能测量的统一方法

• 批准号: 2133452
• 负责人: Ka Pui Mok
• 金额: $ 49.5万
• 依托单位: University of California-San Diego
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-12-15 至 2024-11-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2133452&HistoricalAwards=false
• 关键词: CNS; Core; Small; Unified; Approach

Telecommuting, distance learning, and virtual gatherings have significantly increased during the COVID-19 pandemic. Poor Quality of Experience (QoE) of underlying networked software, such as video conferencing tools, degrades the usability of remote presence applications and thus impairs the effectiveness of distance communication and learning. However, current network performance measurement capabilities cannot systematically capture subjective aspects of QoE, hindering scientific study of this issue.The goal of this project is to develop and evaluate a fundamentally new approach to Internet performance measurement by conducting subjective assessments to measure the QoE of video streaming and conferencing applications in the wild and correlate these QoE measurements with Internet performance. Specifically, this research will 1) develop a unified assessment platform to crowdsource novel experiments for measuring the QoE of video streaming and video conferencing; 2) leverage gamification techniques to incentivize subject participation and improve overall efficiency of the experimental effort; and 3) combine analyses of QoE assessment data with external network topology and performance measurements to diagnose QoE degradations induced by congestion events on interconnections and last-mile access links.This research will advance the field's capabilities to diagnose performance degradation of essential tools for remote learning and telecommuting in the United States. In the process, this project will mentor a diverse set of undergraduate/high school students, who will learn about real-world Internet measurement through participating in platform development.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.

在2019冠状病毒病大流行期间，远程办公、远程学习和虚拟聚会显著增加。底层网络软件（如视频会议工具）的低体验质量（QoE）降低了远程呈现应用程序的可用性，从而损害了远程通信和学习的有效性。然而，目前的网络性能测量能力不能系统地捕捉QoE的主观方面，阻碍了对这一问题的科学研究。该项目的目标是开发和评估一种全新的互联网性能测量方法，通过进行主观评估来测量视频流和会议应用程序的QoE，并将这些QoE测量与互联网性能相关联。具体而言，本研究将1)开发一个统一的评估平台，以众包新的实验来测量视频流和视频会议的QoE；2)利用游戏化技术激励受试者参与，提高实验工作的整体效率；3)将QoE评估数据分析与外部网络拓扑和性能测量相结合，诊断互连和最后一英里接入链路上拥塞事件导致的QoE降级。这项研究将提高该领域诊断美国远程学习和远程办公基本工具性能退化的能力。在此过程中，该项目将指导各种各样的本科生/高中生，他们将通过参与平台开发来学习现实世界的互联网测量。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。