This is the authors' abstract. We don't add key points for this paper.
As Internet of Things (IoT) applications continue to expand, the increasing traffic in frequency spectrum highlights critical challenges related to spectrum utilization, IoT device management, and potential security risk from undetected devices. As such, a comprehensive visibility of IoT devices across the frequency spectrum is essential, especially given the ubiquity of these devices. While terrestrial networks provide local visibility, broader coverage can be achieved through spectrum sensing via non-terrestrial platforms, such as satellites. However, the complex nature of the satellite radio channel introduces additional signal losses, making traditional spectrum sensing techniques inadequate. As such, this work presents a novel methodology for detecting low power radio frames transmitted over a complex satellite channel. We implement joint machine learning comprising a supervised Semantic Segmentation neural network and unsupervised Density-Based Spatial Clustering of Applications with Noise (DBSCAN) to detect IoT packets from the spectrogram images. The proposed method achieves 92% mean accuracy and 93% mean Intersection-over-Union (IoU) at low Signal-to-Noise Ratio (SNR) level. Moreover, the proposed method outperforms the traditional threshold detection methodologies as demonstrated by the Receiver Operating Characteristics (ROC).
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Manzoor et al. (2024) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: