NeTS: EAGER: Exploring Smart Media Pricing In QoE-Driven Network Economics To Revitalize Wireless Multimedia Resource Allocation

全球互联网流量正以惊人的速度急剧增长，对现代网络协议设计提出了重大挑战。此外，多媒体已成为贯穿无线接入网络最后一公里的“最大的大数据”，如视频点播、基于云的视频游戏、社交媒体视频分享等。本项目探索在资源受限的无线接入网络上支持如此大量多媒体流量的潜在解决方案。特别地，本项目通过运用高层次的经济理论，编排低层次的通信资源，以及探索邻近设备之间的自私用户行为，研究提高多媒体用户体验质量（QoE）的新方法。本项目的目标是测试无线大数据多媒体通信挑战的假设，即利用智能媒体定价的新概念，将无线资源分配从费率扭曲方式转变为价格扭曲方式，可以部分解决无线大数据多媒体通信挑战。本项目在传统资源分配之外追求两个连贯的重点：(1)创建一个非统一的质量驱动的智能多媒体定价模型，为具有不同质量贡献和解码影响的多媒体分组分配不同的价格，从而将价格作为资源分配的新维度；(2)设计一个利润驱动的缓存委托协议，通过卸载来提高多媒体质量，从而鼓励用户从邻近设备上清除最近缓存的媒体内容。通过本项目开发的模型、算法和协议可能为无线多媒体资源分配研究的定价策略设计开辟新的前沿。研究成果将被整合到教育课程和学生咨询中，以拓宽学生未来的职业潜力，并为美国的劳动力准备做出贡献。

The global internet traffic is aggressively increasing at an enormous rate, posing significant challenges to modern network protocol design. In addition, multimedia has become the "biggest big data" running through the last mile of wireless access networks, such as video on demand, cloud-based video games, and social media video sharing. This project explores potential solutions to support such big volumes of multimedia traffic over resource-constrained wireless access networks. In particular, this project investigates new approaches to improve quality of experience (QoE) for multimedia users by employing high level economic theories, orchestrating low level communication resources, and exploring selfish user behaviors among neighboring devices.The goal of this project is to test the hypothesis that the wireless big data multimedia communication challenge could be partially solved by leveraging a new concept of smart media pricing to transform the wireless resource allocation from rate-distortion fashion to price-distortion fashion. This project pursues two coherent thrusts beyond traditional resource allocation: (1) creating a non-uniform quality-driven smart multimedia pricing model to assign different prices to multimedia packet groups with different quality contributions and decoding influences, and thus leveraging price as a new dimension in resource allocation; (2) designing a profit-driven cache delegation protocol to improve multimedia quality through offloading, and thus encouraging users to scavenge recently cached media contents from neighboring devices. The models, algorithms and protocols developed through this project could possibly identify new frontiers of pricing policy design in wireless multimedia resource allocation research. Research findings will be integrated to education curriculum and student advising, to broaden students' future career potentials and contribute to the U.S. workforce preparation.