CRII: NeTS: Beyond PHY and Chanel Measurements in Millimeter Wave: Towards Low-Overhead and Resilient Multi-hop Networking

5g及以上技术将使具有超低延迟和极高数据速率要求的应用（如移动AR/VR应用）实现前所未有的激增。无线设备的数量也呈不断增长的趋势，目前已经超过86亿，预计到2022年将增长到123亿。具有高数据速率要求的无线设备密度的增加导致了6ghz以下频段的频谱紧张。30GHz至300GHz之间的毫米波（mmWave）频率可以缓解频谱短缺，并为未来的蜂窝和新兴的数据速率为Gbps的WiFi网络提供主要潜力。目前的研究主要集中在了解无线信道的高频损伤和传播特性。虽然这些努力是必不可少的，但在开发针对毫米波的网络协议方面仍存在差距。该项目采用系统级方法开发和实现可靠的毫米波网络算法，保证端到端性能。它还通过介绍未来无线网络的新趋势和挑战，培训不同层次的弱势群体的学生，并通过吸引人的毫米波应用演示吸引包括高中生在内的年轻人到STEM领域，为课程改进提供了巨大的机会。毫米波和sub-6GHz之间的显著差异要求对协议栈所有层的设计原则进行彻底的重新思考。目前，协议栈的上层在很大程度上仍未被探索，现有的协议也不是为毫米波通信量身定制的。该项目开发协议并实现了具有移动性和链路阻塞的低开销和弹性多跳毫米波通信的测试平台。提出的研究将在三个相互关联的推力上进行：(1)多跳设置的低开销波束对准：利用多臂土方（MAB）框架，该推力将开发一种有效的波束对准算法，该算法将大大减少角波束搜索空间和总体波束对准开销，这对于具有多跳拓扑的密集毫米波网络至关重要；(2)抗阻塞容错多跳路由：为了保证阻塞情况下毫米波通信的可靠性和鲁棒性，该推力将借助网络和路由层解决方案，开发一种能够在链路阻塞情况下快速恢复的按需路由协议；(3)多跳网络中的最优缓冲区分配：本课题将研究多跳毫米波网络中多用户的最优缓冲区分配问题。目标是在延迟和吞吐量性能之间实现最佳权衡，并在不同的数据流之间提供公平的分配。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。

The 5G-and-beyond technologies will enable an unprecedented proliferation of applications with ultra-low latency and extremely-high data rate requirements such as mobile AR/VR applications. There is also an ever-increasing trend in the number of wireless devices that is now already over 8.6 billion and is expected to grow to 12.3 billion by 2022. The increasing density of wireless devices with high data rate requirements has caused spectrum crunch in the sub-6GHz bands. Millimeter-wave (mmWave) frequencies between 30GHz to 300GHz can alleviate the spectrum scarcity and provide major potentials for future cellular and emerging WiFi networks with Gbps data rates. Current studies have mostly focused on understanding wireless channel impairments and propagation characteristics at high frequencies. While such efforts are essential, there are gaps in developing network protocols tailored for mmWave. This project takes a system-level approach to develop and implement algorithms for reliable mmWave networking with guaranteed end-to-end performance. It also provides great opportunities for curriculum enhancement by introducing new trends and challenges in future wireless networks, training students from under-represented populations at different levels, and attracting young minds, including high-school students, to the STEM areas via appealing demos of mmWave applications. The significant differences between mmWave and sub-6GHz call for a radical rethinking of the design principles across all the layers of the protocol stack. Currently, the upper layers of the protocol stack remain largely unexplored and the existing protocols are not tailored for mmWave communication. This project develops protocols and implements a testbed for low-overhead and resilient multi-hop mmWave communication with mobility and link blockage. The proposed research will be conducted across three inter-related thrusts: (1) Low-overhead Beam Alignment for Multi-hop Settings: Leveraging multi-armed bandit (MAB) frameworks, this thrust will develop an efficient beam alignment algorithm that considerably reduces the angular beam search space and overall beam alignment overhead, which is essential for dense mmWave networks with multi-hop topologies; (2) Fault-Tolerant Multi-hop Routing to Combat Blockage: In order to guarantee reliable and robust mmWave communication under blockage, this thrust will resort to network and routing layer solutions to develop an on-demand routing protocol that is able to quickly recover under a link blockage; (3) Optimal Buffer Allocation in Multi-hop Networks: This thrust will investigate the problem of optimal buffer allocation in multi-hop mmWave networks with multi-users. The goal is to strike an optimal tradeoff between delay and throughput performance and provide a fair allocation across different data flows.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.