Special Session 4:
Machine Learning for Radio Frequency Spectrum Awareness
Recent advances in artificial intelligence and machine learning (AI/ML) across modalities such as images, text, audio, and video have spurred growing interest in applying these techniques to the radio frequency (RF) domain. This interest is driven both by commercial challenges such as increasing spectrum congestion and by the critical role of RF communications in defense and national security applications. However, much of the existing research has focused on directly transferring techniques developed for other modalities into RF applications without fully accounting for the unique characteristics of RF signals and sensing environments.
Scope and topics:
This special session aims to bring together researchers and practitioners working on RF-domain-centric machine learning approaches that advance the state of the art in spectrum sensing and spectrum awareness. The session will highlight new algorithms, datasets, and system-level approaches designed specifically for RF environments, as well as research addressing the security, robustness, and trustworthiness of AI/ML methods operating in contested or dynamic spectral environments. By fostering interdisciplinary discussion between the RF signal processing and machine learning communities, this session seeks to accelerate the development of AI/ML techniques that are tailored to the unique challenges of RF sensing.
We invite submissions on the following, but not limited to, topics:
- RF AI/ML Applications
- Spectrum scanning and prioritization
- Signal detection
- Signal classification and identification
- Specific emitter identification
- Co-channel signal separation
- Multi-antenna and distributed sensing
- Collaborative spectrum sensing
- RF AI/ML Datasets and Representations
- RF data modalities (image-based, IQ-based, and feature-based representations)
- Hybrid or “cyborg” dataset generation combining synthetic and real RF data
- Open datasets, benchmarks, and tools for RF machine learning
- RF AI/ML Security
- Adversarial attacks against RF machine learning systems
- Robustness and adversarial hardening methods
- RF AI/ML Trust and Reliability
- Uncertainty quantification in RF ML models
- Explainability and interpretable RF learning systems
Chairs:
- Chair Emails
- Chair Biographies
Dr. William C. Headley: cheadley@vt.edu
Dr. Stephen Adams: scadams21@vt.edu
Dr. Maymoonah Toubeh:may93@vt.edu
Dr. William C. Headley is the Interim Director for the Spectrum Dominance Division at the Virginia Tech National Security Institute where he has served as a principal or co-principal investigator on a multitude of government and commercial projects totaling over \$25M. Within the division, he primarily oversees the Radio Frequency ML and RF Augmented Reality / Virtual Reality portfolios. Through his courtesy appointment within Virginia Tech’s Electrical and Computer Engineering department, he also serves as a mentor and advisor to both undergraduate and graduate student researchers, providing them with hands-on research opportunities through these projects as well as guiding them towards their degree requirements. Beyond those stated, his research areas of interest include synthetic dataset generation, digital signal processing, RF spectrum sensing, and adversarial AI/ML.
Dr. Stephen Adams is a Research Associate Professor in the Intelligent Systems Division of the Virginia Tech National Security Institute. He received a M.S. in Statistics from UVA in 2010 and a Ph.D. from UVA in Systems Engineering in December 2015. His research focuses on applications of machine learning and artificial intelligence in real-world systems. He has experience developing and implementing numerous types of machine learning and artificial intelligence algorithms. His research interests include feature selection, machine learning with cost, transfer learning, reinforcement learning, and probabilistic modeling of systems. His research has been applied to several domains including activity recognition, prognostics and health management, psychology, cybersecurity, data trustworthiness, natural language processing, and predictive modeling of destination given user geo-information data.
Dr. Maymoonah Toubeh: is a Research Assistant Professor in the Spectrum Dominance Division at the Virginia Tech National Security Institute, with research areas including machine learning, computer vision, and robotics. She has significant experience with applying machine learning in the radio frequency space, colloquially termed Radio Frequency Machine Learning (RFML). She has served as PI, Co-PI, and technical performer on several projects pertaining to the RFML space. In particular, she has worked in the RFML areas of modulation classification, signal detection, RF fingerprinting, and RF dataset creation and curation. More generally, she has experience creating robotics and signal datasets from simulation and hardware, as well as building and testing deep neural networks and other machine learning structures. Dr. Toubeh received her Ph.D. and M.S. in computer engineering from Virginia Tech, and her B. Eng. in computer engineering from the American University of Kuwait.
Technical Committee
- Dr. Scott Kuzdeba
- Dr. Joseph Carmack
- Dr. Jacek Kibilda
- Dr. Nicholas Kaminski
Paper Submission Instructions
All papers will be double-blind reviewed and must present original work.
- CMT Submission Site
- Select the track: Special Session 4: Machine Learning for Radio Frequency Spectrum Awareness
Papers submitted for reviewing should conform to IEEE specifications. Manuscript templates can be downloaded from:
Keydates
- Submission Deadline: June 20, 2026
- Notification of Acceptance: July 10, 2026
Registration
In order for your paper to be published in the proceedings you must register to the conference.
Paper Presentation Instructions
The papers submitted to this track will be presented in person as part of the conference. There is no virtual presentation for this session.
ICMLA'26