This is the authors' abstract. We don't add key points for this paper.
With more and more autonomous deployments of wireless networks, accurate knowledge of the RF environment is becoming indispensable. Various techniques have been developed over the years that can not only assess the RF environment but can also characterize the various radio transmitters (sources) that define the ambient RF environment. Machine learning techniques have shown promise for such characterizations through the development of RF machine learning (RFML) systems delivering autonomous control. Although classical machine learning techniques work well for a large variety of tasks, they have not done as well for RFML. For RFML tasks, deep feature learners with an inherent recurrent structure have been shown to perform well. Even so, the field of RFML is still very young, and a lot needs to be done to bridge the gap between the ML community and the wireless community for RFML to be successfully applied for solving large-scale real-life problems. This article is a step in that direction.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Roy et al. (2019) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: