Radio Frequency Machine Learning (RFML) Laboratory

Abstract

Publicly ReleasableONR technical POC. Dr. Santanu DasIn recent years, the Virginia Tech Hume Center for National Security and Technology (#VT Hume#) has established itself as a leading innovator in advancing Radio Frequency Machine Learning (RFML) research in support of our national defense partners. RFML is the unique application of state-of-the-art deep machine learning concepts to wirelesscommunication and electronic warfare applications. In the past 6 years, VT Hume has been awarded RFML research efforts totaling >$13M in funding from our commercial and government sponsors in key RFML innovation areas such as: signal detection and estimation (including LPI/LPD detection) [1-6], signal format identification (e.g. modulation recognition) [1, 4-8], specific emitter identification [9-11], dynamic spectrum access [12], enhancements and robustness [13-17], distributed RFML [18-19], adversarial machine learning [15-16, 20-25] (including both vulnerability analysis and exploitation), andintelligent rf dataset creation [26-28]. Additionally, in support of VT Hume#s primary mission to #cultivate the next generation of national security leaders by developing and executing curricular, extracurricular, and research opportunities to engage students,# we are currently providing resources and faculty mentorship for multiple graduate students in RFML research towards fulfilling their degree requirements and average 20-30 undergraduate research assistants per semester in RFML-based research. At present, these research efforts have led to five successful Master#s degrees[11, 13-15, 21] and many publications [2, 7, 9-10, 19, 22-25, 28]. Recently, we have expanded our RFML student engagement to a joint multi-college effort with Morehouse College, a HBCU, in which undergraduate students from both universities are collaboratively exploring the unique combination of image recognition and RFML techniques for spectrum sensing applications [6].In 2019 we were the recipient of a partial DURIP award ($200k) to cover the expansion of our data storage capabilities for our RFML programs. This DURIP proposal is a request for continuation funding for state-of-the-art GPU-based deep learning training systems in order to enhance our capabilities for executing on these RFML research opportunities of critical importance to our government sponsors and in order to increase our ability to engage the rapidly growing student base with interest in this critical research area. The requested equipment would enhance our capabilities by allowing investigation of much deeper neural network architectures to tackle much more challengingand sophisticated RFML problems and to train these architectures much faster. Additionally, it would allow for more concurrent student machine learning development then is currently achievable within our lab. We believe the requested DGX-A100 machine learning system and machine learning desktop systems fits perfectly with the stated purpose of the DURIP program, enabling university investment in resources and equipment that no individual program at VT Hume could afford.

Document Details

Document Type
DoD Grant Award
Publication Date
Jan 12, 2023
Source ID
N000142312147

Entities

People

  • William Headley

Organizations

  • Office of Naval Research
  • United States Navy
  • Virginia Tech

Tags

Fields of Study

  • Computer science

Readers

  • Mycotoxin ecology in Amazonian ecosystems.
  • Neural Network Machine Learning.
  • Research Science/Academic Research

Technology Areas

  • AI & ML
  • AI & ML - DoD AI Strategy
  • AI & ML - Neural Networks
  • Microelectronics