CIF: Small: Deep Generative Models for Ultra High-Dimensional Communication

CIF：小型：超高维通信的深度生成模型

• 批准号: 2008710
• 负责人: Jeffrey Andrews
• 金额: $ 16.5万
• 依托单位: University of Texas at Austin
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-07-01 至 2021-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2008710&HistoricalAwards=false
• 关键词: CIF; Small; Deep; Generative; Models

Future wireless communications applications like untethered virtual reality, connected vehicles, and high resolution mobile video content delivery will demand very high bit rates and reduced energy consumption. To fulfill these requirements, future wireless systems will use very high carrier frequencies enabled by deploying massive arrays of very small antennas. Such architectures are called ultra high dimensional (UHD) communication systems. This is a very challenging scenario for wireless communication, since the wireless channel is difficult to learn and the wireless energy must be radiated in a highly directional manner towards the intended receiver. Existing approaches to these problems will not scale to the UHD scenario in terms of complexity, power consumption, or overhead signaling. This exploratory project will address these challenges emerging in modern wireless communication systems. The specific focus of this project is on developing novel unsupervised learning techniques to handle these emerging challenges. In particular, this project will enable accurate and efficient channel estimation techniques, design novel data-driven beamforming schemes, and model the nonlinearities created by quantization. This project will further help in enabling power and bandwidth-efficient UHD communication systems, as well as establishing a new methodology and initializing communication over ultra high carrier frequencies (e.g., above 100 GHz).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.

未来的无线通信应用（例如不受限制的虚拟现实、联网车辆和高分辨率移动视频内容传输）将需要非常高的比特率和降低的能耗。为了满足这些要求，未来的无线系统将通过部署大量非常小的天线阵列来使用非常高的载波频率。此类架构称为超高维（UHD）通信系统。对于无线通信来说，这是一个非常具有挑战性的场景，因为无线信道很难学习，并且无线能量必须以高度定向的方式辐射到预期的接收器。解决这些问题的现有方法在复杂性、功耗或开销信令方面无法扩展到超高清场景。这个探索性项目将解决现代无线通信系统中出现的这些挑战。该项目的具体重点是开发新颖的无监督学习技术来应对这些新出现的挑战。特别是，该项目将实现准确高效的信道估计技术，设计新颖的数据驱动波束形成方案，并对量化产生的非线性进行建模。该项目将进一步帮助实现功率和带宽高效的超高清通信系统，以及建立新的方法和初始化超高载波频率（例如，100 GHz 以上）的通信。该奖项反映了 NSF 的法定使命，并被认为是值得的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来获得支持。