SWIFT-SAT: Software Defined Radio based Emulation of SAT-Terrestrial Network Coexistence in "FR3" Bands

7.125 GHz到24 GHz之间的射频频段被称为频率范围3(FR3)，在下一代蜂窝系统中很有希望。该项目创建了一个空中室内FR3试验台，用于地面和卫星系统共存的实验研究，使用软件定义无线电(SDR)来模拟密集的5G蜂窝网络和卫星设备。FR3试验床是现有试验床的扩展，罗格斯大学WINLAB的COSMOS沙箱是一个开放的资源，其他研究人员可以远程访问。该项目使用FR3试验台来研究支持地面和卫星共存的技术，包括用于无线电资源管理、干扰识别和缓解的机器学习。试验台支持的实验研究是对更广泛开展的建模和分析研究的重要补充，有助于更好地决定如何在新的蜂窝系统、现有的主动发射商业卫星(例如数字广播)和现有的被动观测科学卫星(例如天气预报)之间共享FR3，而不会对卫星造成有害干扰。此外，该项目利用WINLAB和罗格斯大学广泛的扩展能力，在多个层面吸引学生参与。该项目侧重于12.2-12.7 GHz频段地面5G与主动商业卫星之间的频谱共享，以及10-10.5 GHz地面5G与10.6-10.7 GHz频段被动科学卫星之间的相邻频段共存。软件和硬件设计用于模拟目标频段的5G无线电网络、商业卫星收发器和无源辐射计。使用超材料软件定义的波束形成来仿真与非静止轨道固定卫星服务(NGSO-FSS)等卫星的共存，并使用热冷定标和与在轨AMSR-E和AMSR2传感器的数据集进行比较，以提高辐射计干扰实验的可靠性和真实性。(1)设计、验证和部署基于FR3 SDR的外差设备、5G新无线电(NR)和SAT波形的仿真以及共存仿真方法；(2)使用仿真实验开发集中式机器学习算法，用于在地面5G和主动商业卫星之间共享无线电资源；以及(3)使用仿真实验评估辐射计对地面5G干扰的敏感度，并开发用于被动科学卫星干扰识别和缓解的机器学习算法。推力2和推力3的工作成果之一是测量分配给地面5G蜂窝系统的频谱比例，该频谱由于用于减轻对卫星的干扰的策略而“丢失”。这一奖励反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The radio frequency band between 7.125 GHz and 24 GHz, known as Frequency Range 3 (FR3), is promising for next generation cellular systems. This project creates an over-the-air indoor FR3 testbed for experimental studies of coexistence between terrestrial and satellite systems, using software-defined radios (SDR) to emulate dense 5G cellular networks and satellite equipment. The FR3 testbed is an extension of an existing testbed, the COSMOS sandbox at Rutgers University WINLAB, and is an open resource remotely accessible to other researchers. This project uses the FR3 testbed to investigate technologies supporting terrestrial-satellite coexistence, including Machine Learning for radio resource management, interference identification and mitigation. Experimental studies enabled by the testbed are important as a complement to more widely performed modeling and analysis studies, to help make better decisions about how to share FR3 between new cellular systems, existing actively transmitting commercial satellites (e.g. digital broadcast) and existing passively observing scientific satellites (e.g. weather forecasting) without harmful interference to the satellites. Additionally, the project engages students at multiple levels, leveraging the broad outreach capabilities of WINLAB and Rutgers.The project focuses on spectrum sharing between terrestrial 5G and active commercial satellites in the 12.2-12.7 GHz band, and adjacent band coexistence between terrestrial 5G at 10-10.5 GHz and passive scientific satellites in the 10.6-10.7 GHz band. Software and hardware are designed that emulate 5G radio networks, commercial satellite transceivers, and passive radiometers in the targeted frequency bands. Metamaterial software-defined beamforming is used to emulate coexistence with satellites like non-geostationary orbit fixed satellite service (NGSO-FSS), and hot-cold calibration and comparison to datasets from on-orbit AMSR-E and AMSR2 sensors is used to enhance the reliability and realism of radiometer interference experiments. There are three thrusts: (1) design, validate and deploy FR3 SDR-based heterodyne devices, emulation of 5G New Radio (NR) and SAT waveforms, and coexistence emulation methods; (2) use emulation experiments to develop centralized Machine Learning algorithms for integrated radio resource management in sharing between terrestrial 5G and active commercial satellites; and (3) use emulation experiments to assess radiometer sensitivity to terrestrial 5G interference and to develop Machine Learning algorithms for interference identification and mitigation by passive scientific satellites. One outcome of the work in thrust 2 and thrust 3 is a measurement of the fraction of spectrum assigned to the terrestrial 5G cellular system that is "lost" due to the strategies used to mitigate interference to satellites.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.