CAREER: Automating the measurement and management of the radio spectrum for future spectrum-sharing applications

职业：自动化无线电频谱的测量和管理，以适应未来的频谱共享应用

• 批准号: 1845858
• 负责人: Mariya Zheleva
• 金额: $ 51.05万
• 依托单位: SUNY at Albany
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-05-15 至 2025-04-30
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1845858&HistoricalAwards=false
• 关键词: CAREER; Automating; measurement; management; radio

A growing number of domains that drive economic growth and humanity's well-being, including healthcare, emergency management and national defense, hinge on wireless network connectivity. This has created a market potential of $640.9 billion, which cannot be realized as existing networks operate at capacity. Despite this potential, only 8% of the radio spectrum is allocated to wireless communication technologies. This minimal allocation creates artificial scarcity of frequency resources, whereby popular bands are saturated, while others are under-utilized. In response, wireless technologies have begun to incorporate new hardware and software to boost their spectrum efficiency through opportunistic frequency reuse. While promising, this trajectory of innovation cannot be sustained unless we establish a framework for principled spectrum measurement and management that can embrace unforeseen network and sensor capabilities in support of future spectrum policy, policing and technology. This project develops a long-term, integrated program of research, education and outreach to (i) establish a scientific and technological framework for automated spectrum measurement in support of shared-spectrum access, and (ii) to train the next generation of wireless specialists at the intersection of networks, digital communications and machine learning. The project will work closely with industry and standardization efforts to ensure broader adoption.The research will be carried out in three thrusts at the confluence of signal processing, digital communications, machine learning, graph mining, and large-scale measurement. First, it will study the effects of real-world scan imperfections on signal features. Following these insights, the project will contribute algorithms for spectrum cognizance and transmitter fingerprinting from imperfect scans. Second, it will enable application-driven measurement by creating an analytical framework that links sensor properties with data quality and algorithm performance. Third, models and algorithms will be integrated in an open system accessible to the community. This project will lay the foundation for future-proof and application-driven spectrum measurement that leverages wide-band and heterogeneous sensing for end-to-end support of emerging wireless networks. It will contribute data-driven and platform-aware abstractions, models and algorithms that produce domain-informed features for automatic spectrum analytics. The outcomes will bridge the existing gap between measurement capabilities and algorithm requirements while tackling realistic scenarios with rogue or incompletely scanned transmitters. The project will produce a cost-utility framework to explore tradeoffs between sensor price, bandwidth, computation and power; and algorithm accuracy, stability and confidence. The integration of research outcomes into an open system will allow systematic and reproducible research and education, while complying with and informing standardization efforts.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.

越来越多推动经济增长和人类福祉的领域，包括医疗保健、应急管理和国防，都依赖于无线网络连接。这创造了6 409亿美元的市场潜力，但由于现有网络满负荷运行，这一潜力无法实现。尽管有这种潜力，但只有8%的无线电频谱被分配给无线通信技术。这种最小化的分配造成了频率资源的人为稀缺，从而使流行的频带饱和，而其他频带利用不足。作为响应，无线技术已经开始结合新的硬件和软件，以通过机会性频率重用来提高其频谱效率。虽然前景看好，但除非我们建立一个原则性的频谱测量和管理框架，可以包含不可预见的网络和传感器功能，以支持未来的频谱政策，政策和技术，否则这种创新轨迹无法持续。该项目开发了一个长期的综合研究，教育和推广计划，以（i）建立自动频谱测量的科学和技术框架，以支持共享频谱接入，以及（ii）在网络，数字通信和机器学习的交叉点上培养下一代无线专家。该项目将与行业和标准化工作密切合作，以确保更广泛的采用。研究将在信号处理、数字通信、机器学习、图形挖掘和大规模测量的汇合点上进行。首先，它将研究真实世界的扫描缺陷对信号特征的影响。根据这些见解，该项目将有助于频谱识别和发射机指纹从不完美的扫描算法。其次，它将通过创建一个将传感器属性与数据质量和算法性能联系起来的分析框架来实现应用驱动的测量。第三，模型和算法将被集成到一个开放的系统中，供社区使用。该项目将为面向未来和应用驱动的频谱测量奠定基础，该测量利用宽带和异构传感技术，为新兴无线网络提供端到端支持。它将提供数据驱动和平台感知的抽象、模型和算法，为自动频谱分析提供领域信息功能。这些成果将弥合测量能力和算法要求之间的现有差距，同时解决具有流氓或不完全扫描的发射机的现实场景。该项目将产生一个成本效用框架，以探索传感器价格，带宽，计算和功率之间的权衡;以及算法的准确性，稳定性和信心。将研究成果整合到一个开放的系统中将允许系统化和可复制的研究和教育，同时遵守和通知标准化工作。该奖项反映了NSF的法定使命，并被认为值得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来支持。

项目成果

Robust and Efficient Modulation Recognition Based on Local Sequential IQ Features

• DOI: 10.1109/infocom.2019.8737397
• 发表时间: 2019-04
• 期刊: IEEE INFOCOM 2019 - IEEE Conference on Computer Communications
• 作者: Wei Xiong;Petko Bogdanov;M. Zheleva

Evolutionary Optimization of Residual Neural Network Architectures for Modulation Classification

• DOI: 10.1109/tccn.2021.3137519
• 发表时间: 2022-06-01
• 期刊: IEEE TRANSACTIONS ON COGNITIVE COMMUNICATIONS AND NETWORKING
• 影响因子: 8.6
• 作者: Perenda, Erma;Rajendran, Sreeraj;Zheleva, Mariya
• 通讯作者: Zheleva, Mariya

Exploiting Self-Similarity for Under-Determined MIMO Modulation Recognition

• DOI: 10.1109/infocom41043.2020.9155247
• 发表时间: 2020-07
• 期刊: IEEE INFOCOM 2020 - IEEE Conference on Computer Communications
• 作者: Wei Xiong;Lin Zhang;M. McNeil;Petko Bogdanov;M. Zheleva

Radio Dynamic Zones: Motivations, challenges, and opportunities to catalyze spectrum coexistence

无线电动态区：促进频谱共存的动机、挑战和机遇

• DOI: 10.1109/mcom.005.2200389
• 发表时间: 2023
• 期刊: IEEE Communications Magazine
• 影响因子: 11.2
• 作者: Zheleva, Mariya;Anderson, Christopher R.;Aksoy, Mustafa;Johnson, Joel T.;Affinnih, Habib;DePree, Christopher G.
• 通讯作者: DePree, Christopher G.

MODELESS: MODulation rEcognition with LimitEd SuperviSion

• DOI: 10.1109/secon52354.2021.9491617
• 发表时间: 2021-07
• 期刊: 2021 18th Annual IEEE International Conference on Sensing, Communication, and Networking (SECON)
• 作者: Wei Xiong;Petko Bogdanov;M. Zheleva