Collaborative Research: MLWiNS: ANN for Interference Limited Wireless Networks

合作研究：MLWiNS：干扰有限无线网络的 ANN

• 批准号: 2003082
• 负责人: Xiao Fu
• 金额: $ 18.55万
• 依托单位: Oregon State University
• 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: 2020-06-01 至 2024-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2003082&HistoricalAwards=false
• 关键词: Collaborative; Research; MLWiNS; ANN; Interference

Demand for wireless data services will continue to rise rapidly in the foreseeable future. The goal of this project is to develop new advanced solutions for wireless access networks with the objectives of increasing network throughput and maximizing overall network utilities. Long-serving model-based solutions are facing severe limitations due to delays in tracking the ever-changing radio and network environment as well as measurement inaccuracies. The main novelty of this project is to bring new tools based on artificial neural networks (ANN) to meet those challenges. In particular, this project will investigate when, how, and why ANN-based learning techniques can be applied to a wide range of wireless networking problems with realistic constraints. This project will pursue transformative solutions that aim to benefit academia and industry alike. Specifically, this project will marry supervised and unsupervised learning techniques with time-tested models of physical resources, channels, traffic, and network utilities. An important task is to exploit commonalities of ANN-based solutions for a number of subproblems to develop a set of principled, holistic solutions for the overall wireless networking problem, seeking solutions that are scalable, computationally efficient, and highly adaptive. Pertinent learnability and complexity theories backing the solutions will also be developed, in order to offer generalizable design principles. The ANN-based solutions developed are expected to be a major building block of next generation wireless access networks with associated economic benefits.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.

在可预见的未来，对无线数据服务的需求将继续快速增长。该项目的目标是为无线接入网络开发新的高级解决方案，目标是增加网络吞吐量和最大限度地提高整体网络效用。由于跟踪不断变化的无线电和网络环境的延迟以及测量的不准确性，基于长期服务模型的解决方案正面临严重的限制。该项目的主要创新之处在于引入了基于人工神经网络(ANN)的新工具来应对这些挑战。特别是，这个项目将调查基于人工神经网络的学习技术何时、如何以及为什么可以应用于具有现实约束的广泛的无线网络问题。该项目将寻求变革性的解决方案，旨在使学术界和工业界都受益。具体地说，该项目将把有监督和无监督的学习技术与经过时间考验的物理资源、通道、流量和网络公用事业的模型结合起来。一项重要的任务是利用许多子问题的基于人工神经网络的解决方案的共性，为整个无线网络问题开发一套原则性的、整体的解决方案，寻求可扩展、计算高效和高度自适应的解决方案。还将开发支持解决方案的相关可学习性和复杂性理论，以提供可推广的设计原则。开发的基于ANN的解决方案预计将成为下一代无线接入网络的主要组成部分，并带来相关的经济利益。该奖项反映了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。

项目成果

Identifiability-Guaranteed Simplex-Structured Post-Nonlinear Mixture Learning via Autoencoder

• DOI: 10.1109/tsp.2021.3096806
• 发表时间: 2021-06
• 期刊: IEEE Transactions on Signal Processing
• 影响因子: 5.4
• 作者: Qi Lyu;Xiao Fu

Learning to Continuously Optimize Wireless Resource in Episodically Dynamic Environment

• DOI: 10.1109/icassp39728.2021.9413503
• 发表时间: 2020-11
• 期刊: ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
• 作者: Haoran Sun;Wenqiang Pu;Minghe Zhu;Xiao Fu;Tsung-Hui Chang;Mingyi Hong

Deep Spectrum Cartography: Completing Radio Map Tensors Using Learned Neural Models

• DOI: 10.1109/tsp.2022.3145190
• 发表时间: 2021-05
• 期刊: IEEE Transactions on Signal Processing
• 影响因子: 5.4
• 作者: S. Shrestha;Xiao Fu;Min-Fong Hong

Deep Generative Model Learning For Blind Spectrum Cartography with NMF-Based Radio Map Disaggregation

• DOI: 10.1109/icassp39728.2021.9413382
• 发表时间: 2021-06
• 期刊: ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
• 作者: S. Shrestha;Xiao Fu;Min-Fong Hong

Optimal Solutions for Joint Beamforming and Antenna Selection: From Branch and Bound to Graph Neural Imitation Learning

• DOI: 10.1109/tsp.2023.3244096
• 发表时间: 2022-06
• 期刊: IEEE Transactions on Signal Processing
• 影响因子: 5.4
• 作者: S. Shrestha;Xiao Fu;Mingyi Hong