CAREER: Enabling Generative, Robust, and Interpretable Wireless Intelligence for RF Spectrum Foundations and Applications

Artificial intelligence (AI) and machine learning (ML) are envisioned to greatly transform the platform-based ecosystems, business models, and services in next-generation (NextG) wireless systems. Further, wireless infrastructure and electromagnetic spectrum resources are shared to provide both sensing/localization and communication services in NextG wireless networks, thus improving user’s experience and impacting people’s lives. This project aims to develop AI/ML-native wireless intelligence by integrating generative, robust, and interpretable ML methods with wireless systems for radio frequency (RF) sensing, channel modeling, and spectrum access. It also provides rich hands-on learning, research, and training opportunities for both undergraduate and graduate students through the integration of research and education. This project addresses three major challenges in wireless AI/ML: insufficient high-quality data, domain and target shift, and the black-box nature of many ML models. The research is organized into four synergistic thrusts. First, it develops spectrum data synthesis and augmentation methods by exploring recent advances in generative AI models to address the need for sufficient high-quality labeled data. Second, it investigates domain generalization for wireless sensing and spectrum utilization, focusing on integrated foundation models, uncertainty estimation, and domain-invariant wireless feature representations. Third, it improves interpretability of ML-guided wireless systems by developing interpretable ML models for spectrum applications and enhancing the understanding of RF spectrum learning and generalization. Fourth, it integrates, validates, and evaluates the proposed methods through in-depth studies of representative wireless system applications, including RF fingerprinting, wireless sensing, and channel prediction. 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. NSF Award ID: 2543148 | Program: 01002627DB NSF RESEARCH & RELATED ACTIVIT,01003031DB NSF RESEARCH & RELATED ACTIVIT,01002930DB NSF RESEARCH & RELATED ACTIVIT | Principal Investigator: Xuyu Wang | Institution: Florida International University, MIAMI, FL | Award Amount: $323,004 View on NSF Award Search: https://www.nsf.gov/awardsearch/show-award/?AWD_ID=2543148 View on Research.gov: https://www.research.gov/awardapi-service/v1/awards/2543148.html