CAREER: Quality of Experience and Network Management in the Encrypted Internet

虽然加密对安全和隐私至关重要，但它限制了内容提供商（例如Netflix和YouTube）和移动网络运营商（例如at&t和Verizon）有效和高效地向用户提供内容的能力。首先，加密限制了内容提供商保持对现代应用程序（如增强现实或虚拟现实（AR/VR）视频流）有效控制的能力。其次，加密限制了移动网络运营商在其众多用户之间有效共享稀缺无线电频谱的能力。这些问题会显著降低用户的体验质量（QoE）。这个项目的目标是在日益加密的互联网中提高现代应用程序的QoE。该项目将研究网络运营商和内容提供商在加密互联网中为QoE和网络管理进行的隐式和显式合作。首先，该项目将通过基于机器学习的加密流量分析，调查网络运营商和内容提供商之间的隐性合作。其次，该项目将研究用于网络运营商和内容提供商之间显式信息交换的带内和带外协议和API框架。该项目研究了一系列用于合作QoE和网络管理的创新方法，范围从全新的方法到向后兼容和增量部署的方法。该研究将有助于改善AR/VR视频流等现代高吞吐量低延迟应用的互联网体验，特别是对于互联网连接差的服务不足地区的用户。该项目还将吸引本科生和少数族裔学生。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。

While encryption is essential for security and privacy, it limits the ability of content providers (e.g. Netflix and YouTube) and mobile network operators (e.g. AT&T and Verizon) to effectively and efficiently deliver content to users. First, encryption limits the ability of content providers to maintain effective control over modern applications such as augmented reality or virtual reality (AR/VR) video streaming. Second, encryption limits the ability of mobile network operators to efficiently share the scarce radio spectrum among their many users. These issues can significantly degrade users' Quality of Experience (or QoE). The goal of this project is to improve QoE of modern applications in the increasingly encrypted Internet. This project will investigate implicit and explicit cooperation between network operators and content providers for QoE and network management in the encrypted Internet. First, the project will investigate implicit cooperation between network operators and content providers through machine learning based analysis of encrypted traffic. Second, the project will investigate in-band and out-of-band protocols and API frameworks for explicit information exchange between network operators and content providers. The project investigates a spectrum of innovative approaches for cooperative QoE and network management, ranging from clean slate approaches to the ones that are backward compatible and incrementally deployable. The proposed research will help improve Internet experience of modern high-throughput low-delay applications such as AR/VR video streaming, especially for users in underserved areas with poor Internet connectivity. The project will also engage undergraduates and under-represented minority students.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.