Collaborative Research: CNS Core: Small: Beyond-5G Extreme Mobility: Issues and Solutions

合作研究：CNS 核心：小型：超越 5G 极限移动性：问题与解决方案

• 批准号: 2008824
• 负责人: Hyeji Kim
• 金额: $ 25万
• 依托单位: University of Texas at Austin
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-10-01 至 2024-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2008824&HistoricalAwards=false
• 关键词: Collaborative; Research; CNS; Core; Small

The current 4G/5G cannot ensure satisfactory reliability and performance for many emerging usage scenarios, such as vehicle-to-everything, high-speed rails, low earth orbit satellites, and drones. This project proposes a forward-looking solution suite for beyond-5G extreme mobility. It eliminates the 4G/5G limitations for extreme mobility by enabling predictive wireless design and simplified yet more efficient wide-area mobility support. If successful, this work will significantly advance the state-of-the-art in wireless networks. To maximize the impact, the researchers will collaborate with industry and release software to the public. The research outcome will also be incorporated into the graduate and undergraduate curriculum.This project proposes a forward-looking, transformative solution suite to beyond-5G extreme mobility. It first unveils 5G's deficiencies in extreme mobility under various scenarios and studies the client and infrastructure's proper roles in supporting them. It then develops novel approaches to accurately predict wireless channel by extracting delay-Doppler of the multipath and exploiting their temporal locality. In this way, it eliminates slow channel feedback in high-speed mobility and adapts to the fast-varying channel in extreme mobility. It further uses the predicted channel to enable predictive rate adaptation, resource allocation, and MIMO optimization under high mobility. Finally, it designs latency-friendly and interpretable distributed machine learning (ML) to help clients analyze the latency bottlenecks, and perform cross-layer latency optimizations. The proposed research will be evaluated using a software-defined radio prototype and large-scale emulation driven by operational 4G/5G traces.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.

目前的4G/5G无法确保许多新兴使用场景的令人满意的可靠性和性能，例如车辆到一切，高速铁路，低地球轨道卫星和无人机。该项目为超越5G的极端移动性提出了一个前瞻性的解决方案套件。它通过实现预测性无线设计和简化但更高效的广域移动支持，消除了4G/5G对极端移动性的限制。如果成功，这项工作将大大推进无线网络的最新发展。为了最大限度地发挥影响力，研究人员将与行业合作，并向公众发布软件。研究成果也将被纳入研究生和本科生课程。该项目提出了一个前瞻性的，变革性的解决方案套件，以超越5G的极端移动性。它首先揭示了5G在各种场景下极端移动性的不足，并研究了客户端和基础设施在支持它们方面的适当作用。然后，它开发了新的方法来准确地预测无线信道提取延迟多普勒的多径和利用其时间局部性。通过这种方式，它消除了高速移动性中的慢信道反馈，并适应极端移动性中的快变化信道。它还使用预测的信道来实现高移动性下的预测速率自适应、资源分配和MIMO优化。最后，它设计了延迟友好和可解释的分布式机器学习（ML），以帮助客户端分析延迟瓶颈，并执行跨层延迟优化。该奖项反映了NSF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。

项目成果

Inventing Codes for Channels With Active Feedback via Deep Learning

• DOI: 10.1109/jsait.2022.3216515
• 发表时间: 2022-09
• 期刊: IEEE Journal on Selected Areas in Information Theory
• 作者: Karl Chahine;Rajesh K. Mishra;Hyeji Kim

Movement-Based Reliable Mobility Management for Beyond 5G Cellular Networks

• DOI: 10.1109/tnet.2022.3190788
• 发表时间: 2023-02
• 期刊: IEEE/ACM Transactions on Networking
• 作者: Zhehui Zhang;Yuanjie Li;Qianru Li;Jinghao Zhao;Ghufran Baig;L. Qiu;Songwu Lu

Learning Variable-Rate Codes for CSI Feedback

• DOI: 10.1109/globecom48099.2022.10000622
• 发表时间: 2022-12
• 期刊: GLOBECOM 2022 - 2022 IEEE Global Communications Conference
• 作者: Heasung Kim;Hyeji Kim;G. Veciana