RINGS: Ensuring Reliability in mmWave Networks

该项目旨在为多跳毫米波网络设计弹性低复杂度的传输机制，以实现可证明的接近最佳性能并提供服务保证。在无线领域，几个重要的用例依赖于多跳毫米波网络部署，包括购物中心、机场、博物馆和企业等专用网络；在密集的城市场景中使用毫米波链路作为回程的毫米波网状网络；采用移动热点的军事应用；以及基于毫米波的车联网（V2X）服务，如协同感知。尽管前面提到了毫米波通信的前景，但众所周知，由于毫米波频谱的敌对传播性质，毫米波链路对阻塞非常敏感，信道可能突然改变，路径被中断。此外，与传统无线网络一样，由于自然灾害和资源枯竭等因素，毫米波网络节点容易受到组件故障的影响。如果成功，该项目旨在提供能够抵御此类中断的新型传输机制。为了实现研究目标，该项目建立在所谓的1-2-1网络模型之上，该模型为毫米波网络提供了一个简单但信息丰富的模型。项目工作包括设计主动机制，旨在通过利用从多路径多样性到多层次和网络编码等工具提前建立弹性；设计反应机制，识别和适应链接或节点故障，使用反压力算法和强化学习等工具；并综合上述机制，使一套工具能够服务于各种应用程序。区分毫米波网络与图形表示网络的一个关键特征是波束调度，它既提供了建立弹性的机会，也提供了主要在操作复杂性方面的挑战。该项目开发的解决方案汇集了来自多个学科的工具；特别是，它结合了信息论、机器学习、算法和网络的工具来解决所提出的目标。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。

This project aims to design resilient low-complexity transmission mechanisms for multi-hop mmWave networks that achieve provably near-optimal performance and provide service assurances. Within the wireless landscape, several important use cases rely on multihop mmWave network deployments, including private networks, such as in shopping centers, airports, museums and enterprises; mmWave mesh networks that use mmWave links as backhaul in dense urban scenarios; military applications employing mobile hotspots; and mmWave based vehicle-to-everything (V2X) services, such as cooperative perception. Despite the aforementioned promising aspects of mmWave communication, it is well known that -- because of the hostile propagation nature of the mmWave spectrum -- mmWave links are highly sensitive to blockage, channels may abruptly change and paths get disrupted. Moreover, as in traditional wireless networks, mmWave network nodes are susceptible to component failures, due to factors such as natural hazards and resource depletion. If successful, this project aims to provide novel transmission mechanisms that are resilient against such disruptions.Towards achieving the objectives of the research, the project builds on the so-called 1-2-1 network model that offers a simple yet informative model for mmWave networks. The project work includes the design of proactive mechanisms, that aim to build resilience in advance by exploiting tools that range from multipath diversity to multilevel and network coding; the design of reactive mechanisms, that identify and adapt to link or node failures, using tools such as backpressure algorithms and reinforcement learning; and synthesis of the above mechanisms to enable a suite of tools that can serve a variety of applications. A key feature that differentiates mmWave networks from networks that are represented as graphs is beam scheduling, which offers both opportunities to build resilience, as well as challenges mainly in terms of complexity of operation. The project develops solutions that bring together tools from several disciplines; in particular, it combines tools from information theory, machine learning, algorithms, and networking to address the proposed goals.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.