User Tools
rings
Differences
This shows you the differences between two versions of the page.
| Both sides previous revision Previous revision Next revision | Previous revision | ||
|
rings [2023/07/20 14:55] lingjialiu |
rings [2023/07/21 14:51] (current) lingjialiu |
||
|---|---|---|---|
| Line 14: | Line 14: | ||
| ===== Personnel ===== | ===== Personnel ===== | ||
| - | * PI in the Leading Institution: [[https://computing.ece.vt.edu/~lingjialiu/doku.php|Lingjia Liu]] (ECE at VT) | + | * [[https://computing.ece.vt.edu/~lingjialiu/doku.php|Lingjia Liu]], ECE at VT: PI in the Leading Institution |
| - | * PI in the Collaborative Institution: [[https://ece.duke.edu/faculty/robert-calderbank|Robert Calderbank]] (ECE/CS at Duke) | + | * [[https://ece.duke.edu/faculty/robert-calderbank|Robert Calderbank]], ECE/CS at Duke: PI in the Collaborative Institution |
| - | * PI in the Collaborative Institution: [[https://users.ece.cmu.edu/~yuejiec/|Yuejie Chi]] (ECE at CMU) | + | * [[https://users.ece.cmu.edu/~yuejiec/|Yuejie Chi]], ECE at CMU: PI in the Collaborative Institution |
| + | * Shadab Mahboob, ECE at VT: PhD student | ||
| + | * Beyza Dabak, ECE at Duke: PhD student | ||
| + | * Shicong Cen, ECE at CMU: PhD Student | ||
| + | ===== Industry Collaborators ===== | ||
| + | * Santhosh Kumar Vanaparthy | ||
| + | * Jan Schreck | ||
| + | * Bin Li | ||
| + | * Zhu Zhou | ||
| + | * Jerry Syder | ||
| - | ===== Abstract ===== | + | ===== Synopsis ===== |
| ### | ### | ||
| - | With the growing importance of wireless connectivity for social and economic interaction, there have been rising demands for greater spectrum use by both primary and secondary active radios, amidst the critical requirement of quiet spectrum for scientific exploration by receiver-only passive systems. To improve the overall spectrum utilization of the wireless ecosystem, this project develops a holistic Intelligent Dynamic spEctrum Access (IDEA) framework that can substantially enhance the spectrum utilization, energy-efficiency, and coexistence capability of spectrum sharing networks. In IDEA, enabling technical innovations across multiple disciplines are synergistically developed, including neuromorphic design of energy-efficient computing hardware at the device and circuit level, and artificial intelligence for spectrum sensing and dynamic access at the network level. The spectrum and interference management in IDEA conscientiously treats the coexistence constraints imposed by passive services, in support of scientific and societal returns from remote sensing investments. The outcomes of this research are expected to broadly impact next-generation wireless networks and Internet of Things applications with high traffic demands, such as autonomous driving, smart cities and remote sensing.\\ | + | The explosive growth of mobile data traffic is in part a response to the proliferation of mobile access services in recent years. However, not all mobile users are able to enjoy stable and reliable broadband connections due to limited network capacities and limited coverage areas. Therefore, in next generation (NextG) mobile broadband networks it is necessary to integrate terrestrial and non-terrestrial networks to democratize wireless access, by providing seamless wireless coverage and supporting heterogeneous service requirements. To meet this goal, this project will develop the fundamental research necessary to integrate and operate terrestrial and non-terrestrial networks, termed Ground and Air Integrated Networks (GAINs). The research project is highly interdisciplinary at the interface of machine learning and wireless networks, providing graduate and undergraduate students with the skills needed to thrive in either community, as well as to bridge them either in academia or in industry. Software and hardware testbeds will provide proof of concept demonstrations for academic, industry and government partners.\\ |
| \\ | \\ | ||
| - | The goal of this project is to develop an intelligent dynamic spectrum access framework with unprecedented spectrum utilization efficiency and agility to support spectrum coexistence. The developed IDEA network platform supports heterogeneous devices from both primary and secondary active radios as well as passive radios. Key technical innovations are developed across the network to substantially enhance system-level spectrum utilization and active-passive radio coexistence. Specifically, analog/mixed-signal spiking neural network (SNN)-based neuromorphic computing hardware is designed to provide on-board intelligence at ultra-low power for resource-constrained secondary active radios. Model-free deep reinforcement learning is integrated with wireless domain knowledge and the SNN platform to accelerate learning-based spectrum access and coexistence. Advanced spectrum monitoring techniques are developed to quickly detect and characterize various signal emitters in both active and passive services. Finally, software and hardware testbeds are developed for system level evaluation and tradeoff optimization. The IDEA framework not only empowers efficient spectrum access in highly dynamic wireless environments, but also facilitates holistic system design and optimization across devices and circuits, sensing and communications, and networking. | + | The overarching objective of this research program is to develop fundamental enabling communication and computing technologies for resilient and intelligent Ground and Air Integrated Networks (GAINs) based on waveform design, real-time machine learning, resource scheduling, distributed computing and learning. This research program makes the sparse representation of the propagation environment visible to machine learning algorithms by designing signals and controlling networks in the delay-Doppler domain, rather than the time-frequency domain. The research program is streamlined into four interconnected research thrusts: 1) Waveform design to enable machine learning; 2) multi-agent reinforcement learning-enabled resilient scheduling for terrestrial networks; 3) distributed and resilient computing in GAINs; and 4) proof-of-concept development and system evaluation. A new suite of distributed and resilient machine learning algorithms that are communication-efficient and heterogeneity-aware will be tailored to information processing in GAINs at the speed of the next generation (NextG) networks. |
| ### | ### | ||
| ===== Research Breakdown ===== | ===== Research Breakdown ===== | ||
| - | **IDEA** covers the following research components: | ||
| - | * Analog/mixed-signal neuromorphic computing hardware: SNN-aided device designs including multiplexing neural encoding, computing-in-memory, and efficient training for resource constrained secondary radios to enable on-board intelligence at ultra-low power consumption and compact design areas; | ||
| - | * Improving spectrum utilization and coexistence through learning: tailored integration of model-free DRL and domain knowledge of spectrum sharing network with improved sample efficiency to increase spectrum utilization in realistic scenarios, along with judiciously designed DSA actions for coexistence; | ||
| - | * Spectrum sensing through concise statistical modeling and learning: efficient spectrum sensing techniques that exploit the inherent structural information of statistics to accurately extract discriminative higher-order statistical features of various signal sources within a short sensing time; | ||
| The following thrusts are organized to address the above mentioned research components: | The following thrusts are organized to address the above mentioned research components: | ||
| - | * Thrust 1: Energy-Efficient Spiking Neural Networks Design and Optimization; | + | * Thrust 1: Waveform Design to Enable Machine Learning ---- Viewing Channels from the Delay-Doppler Domain |
| - | * Thrust 2: Accelerating Learning-based DRL to Improve Spectrum Utilization; | + | * Thrust 2: Multi-Agent Reinforcement Learning-Enabled Resilient Scheduling for TNs ---- Reduced Scheduling Complexity with Inaccurate Input Information |
| - | * Thrust 3: Spectrum Sensing and Interference Control for Active and Passive Radio Coexistence. | + | * Thrust 3: Distributed and Resilient Computing in GAINs ---- Focus on Heterogeneity and Decentralized Nature |
| + | ### | ||
| + | Designing waveforms in a different domain (e.g., the delay-Doppler domain) creates a channel that only changes at the speed of the local propagation environment, enabling detection, estimation and learning at the speed of NextG, and migration of MIMO scheduling to the cloud. Reservoir computing-based real-time machine learning tools will be coupled with new waveform design to enable receive processing at the speed of NextG. Novel multi-agent reinforcement learning-based centralized scheduling strategies will be introduced to enable low-complexity and resilient resource allocation in the delay-Doppler domain. A new suite of distributed and resilient machine learning algorithms that are communication-efficient and heterogeneity-aware will be tailored to information processing in GAINs. | ||
| + | ### | ||
| + | |||
| + | ===== Education and Outreach Activities ===== | ||
| + | |||
| + | * PI Liu, WiOPT Workshop on the Machine Learning in Wireless Communication Networks (WMLC), Real-Time Machine Learning for MIMO-OFDM: Symbol Detection Using Reservoir Computing, September 2022 | ||
| + | * PI Liu, 4th Buffalo Day for 5G and Wireless Internet of Things, Real-Time Machine Learning for MIMO-OFDM: Symbol Detection Using Reservoir Computing, November 2022 | ||
| + | * PI Calderbank, NextG Alliance, Online Seminar: Learning in the Delay Doppler Domain, September 2022 | ||
| + | * PI Calderbank, USC Viterbi School of Engineering, The Viterbi Lecture: Learning to Communicate, March 2023 | ||
| + | * PI Calderbank, University of Arizona: Learning to Communicate, March 2023 | ||
| + | * PI Calderbank, Apple Online Seminar: Learning to Communicate, April 2023 | ||
| + | * PI Calderbank, White House 6G Wireless Forum at the National Science Foundation: Panelist, Federal Funding Panel (moderated by Margaret Martonosi) with Peter Vetter (President, Bell Labs Core Research) | ||
| + | * PI Calderbank, UCLA: Learning to Communicate, May 2023 | ||
| + | * PI Calderbank, Nokia Bell Labs, Online Seminar: Learning to Communicate, May 2023 | ||
| + | * PI Calderbank, Middle East Technical University (METU) Online Seminar: Learning to Communicate, May 2023 | ||
| + | * PI Chi, CVPR Workshop on Federated Learning for Computer Vision (FedVision): Coping with Heterogeneity and Privacy in Communication-Efficient Federated Optimization, June 2022 | ||
| + | * PI Chi, Lehigh University: Coping with Heterogeneity and Privacy in Communication-Efficient Federated Optimization, October 2022 | ||
| + | * PI Chi, Cornell University: Multi-agent Reinforcement Learning: Statistical and Optimization Perspectives, November 2022 | ||
| + | * PI Chi, IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) Tutorial: Advances in Federated Optimization: Efficiency, Resiliency, and Privacy, June 2023 | ||
| ===== Publication ===== | ===== Publication ===== | ||
| + | - Cen, Shicong and Chen, Fan and Chi, Yuejie "Independent Natural Policy Gradient Methods for Potential Games: Finite-time Global Convergence with Entropy Regularization" IEEE Conference on Decision and Control (CDC) , 2022 | ||
| + | - Zhan, Wenhao and Cen, Shicong and Huang, Baihe and Chen, Yuxin and Lee, Jason D. and Chi, Yuejie "Policy Mirror Descent for Regularized Reinforcement Learning: A Generalized Framework with Linear Convergence" SIAM Journal on Optimization , v.33 , 2023 | ||
| + | - Li, Z. and Zhao, H. and Li, B. and Chi, Y. "SoteriaFL: A Unified Framework for Private Federated Learning with Communication Compression" Advances in neural information processing systems , 2022 | ||
| + | - Zhao, H. and Li, B. and Li, Z. and Richtarik, P. and Chi, Y. "BEER: Fast O(1/T) Rate for Decentralized Nonconvex Optimization with Communication Compression" Advances in neural information processing systems , 2022 | ||
| + | - Ao, R. and Cen, S. and Chi, Y. "Asynchronous Gradient Play in Zero-Sum Multi-agent Games" International Conference on Learning Representations (ICLR) , 2023 | ||
| + | - Cen, S. and Chi, Y. and Du, S. and Xiao, L. "Faster Last-iterate Convergence of Policy Optimization in Zero-Sum Markov Games" International Conference on Learning Representations (ICLR) , 2023 | ||
| + | |||
| + | ===== Code Repository ===== | ||
rings.1689879335.txt.gz · Last modified: 2023/07/20 14:55 by lingjialiu
Page Tools
Except where otherwise noted, content on this wiki is licensed under the following license: CC Attribution-Share Alike 4.0 International
