NeTS: Small: Enabling Mobile mmWave Communication: Achieving Low Power and Delay via a Hybrid RF Design

到2020年，年数据流量预计将超过130艾比特。这种流量洪流将显著加剧蜂窝运营商已经经历的频谱紧缩问题。为了支持流量需求的显著增长，预计5G蜂窝系统将使用毫米波(MmWave)频段的某些部分，跨越30 GHz至300 GHz的频谱。这将大大增加蜂窝提供商的可用频谱，这些提供商目前仅限于700 MHz到2.6 GHz之间的频谱，为所有当前的蜂窝技术只分配了780 MHz的带宽。然而，在毫米波通信成为现实之前，由于需要解决的通道可变、间歇性连接、高延迟和高能量使用等问题，它面临着巨大的挑战。为了克服这些挑战，该项目将开发和实施必要的架构和相关算法，以完全集成的方式利用毫米波和射频接口。这项拟议的研究是雄心勃勃的，但如果成功，将有望朝着移动毫米波技术的发展迈出重要的一步，这将显著缓解我们目前面临的巨大频谱紧缩。拟议的研究将跨越三个相互关联的主要推动力：1.移动毫米波传播的缩小实验：实验证明，当前无线传播模型所做的时间尺度分离假设不适用于移动毫米波。该项目将开发一种测量装置，该装置针对不同感兴趣的环境使用小比例模型，并按比例高频进行测量。这一新颖的框架能够在射频和毫米波接口上联合进行准确的移动测量，而不必构建实际的真实系统。节能通信：尽管毫米波接口提供了巨大的带宽，但始终充分利用它并不节能。此外，连接建立和波束成形可能会消耗大量资源。为了缓解这些挑战，本项目将仔细利用射频和毫米波接口之间的相关性，以最大限度地提高每单位功耗的实现速率。主要目标之一是完全在模拟领域处理快速毫米波波束形成。低延迟通信：考虑到毫米波信道的高度变化性，该项目将开发一个框架来联合管理RF和毫米波接口的队列。相关的延迟最小化和队列管理方案明确地考虑了相对于毫米波链路速率的处理器读/写速度。

By 2020, annual data traffic is expected to surpass 130 exabits. This deluge of traffic will significantly exacerbate the spectrum crunch problem that cellular providers are already experiencing. To support this significant increase in traffic demand, it is envisioned that 5G cellular systems will use certain portions of the millimeter-Wave (mmWave) band, spanning the spectrum between 30GHz to 300GHz. This will substantially increase the spectrum available for cellular providers who are currently confined to the frequency spectrum between 700MHz and 2.6GHz with only 780MHz of bandwidth allocation for all current cellular technologies. However, before mmWave communication can become a reality, it faces significant challenges due to variable channels, intermittent connectivity, high delay, and high-energy usage that need to be addressed. To overcome these challenges, this project will develop and implement the necessary architecture and associated algorithms to exploit the mmWave and RF interfaces in a fully integrated fashion. The proposed research is ambitious, but if successful is expected to take an important step towards the development of mobile mmWave technology, which will significantly alleviate the enormous spectrum crunch that we are currently encountering.The proposed research will be conducted across three major inter-related thrusts:1. Scaled-down experimentation for mobile mmWave propagation: It has been experimentally demonstrated that the time-scale separation assumption made by the current models of wireless propagation does not work for mobile mmWave. This project will develop a measurement setup that uses small-scale models for different environments of interest and take the measurements at proportionally high frequencies. This novel framework enables accurate mobile measurements, jointly at the RF and mmWave interfaces without having to build actual real-world systems.2. Energy-efficient communication: Despite the availability of huge bandwidths at the mmWave interface, it is not energy efficient to utilize it fully at all times. Furthermore, connection setup and beamforming can consume significant resources. To alleviate these challenges, this project will carefully exploit the correlations across RF and mmWave interfaces in order to maximize the achieved rate per unit power consumed. One of major goals is to handle fast mmWave beamforming completely in the analog domain.3. Low-delay communication: Taking into account the high variability of mmWave channels, this project will develop a framework to jointly manage the queues at the RF and mmWave interfaces. The associated delay minimization and queue management schemes explicitly take into account the processor read/write speed relative to the mmWave link rates.