Learning-Enabled RF Signal Processing for Intelligent Wireless Systems

Wireless systems increasingly operate in complex, dynamic, and heterogeneous radio-frequency environments, where interference, hardware impairments, rapidly varying channels, and dense spectrum use challenge conventional signal-processing pipelines. At the same time, modernradios expose rich data sources, including raw I/Q samples, channel state information, spectrum measurements, and distributed sensing observations. Machine learning offers new tools for extracting structure and adapting to changing conditions, but practical wireless deployment requires methods that respect RF physics, signal structure, and stringent latency, energy, data, and computation constraints. These developments motivate learning-enabled RF signal processing that combines data-driven adaptation with the robustness, efficiency, and interpretability required by intelligent wireless systems.This Research Topic aims to bring together advances at the intersection of machine learning, RF signal processing, and wireless communications. The central challenge is to develop learning-enabled methods that improve detection, classification, spectrum awareness, sensing, localization, and inference while remaining reliable under realistic wireless conditions and deployable on resource-constrained platforms. We seek approaches that exploit signal and channel structure rather than treating RF measurements as generic data, and that address limited labels, changing propagation environments, interference, hardware nonidealities, distribution shift, communication constraints, and computational budgets. Of particular interest are methods that connect algorithmic advances to operational wireless objectives and demonstrate robustness, efficiency, and generalization in realistic simulations, testbeds, or over-the-air experiments.We welcome Original Research, Methods, Reviews, and Perspective articles on RF signal detection and classification; spectrum sensing and interference identification; RF sensing, localization, and tracking; learning from I/Q samples, channel state information, and other radiomeasurements; model-based or physics-informed learning for communications; distributed and edge RF processing; federated learning for wireless sensing; robust and compute-efficient inference; learning under channel variation, interference, hardware impairments, and distributionshift; self-supervised and data-efficient RF learning; adaptive receiver and signal-processing design; and datasets, benchmarks, testbeds, and over-the-air evaluations. Contributions should maintain a clear connection to RF signal processing and emphasize methodological, systemlevel, or experimental advances for intelligent wireless systems. Topic editors: Erdem Koyuncu, Kevin Chan, Peter Chong Source: https://www.frontiersin.org/research-topics/86391/learning-enabled-rf-signal-processing-for-intelligent-wireless-systems