A Hybrid Collaborative Time–Frequency Network for NOMA Spectrum Sensing
IEEE Communications Letters · 2026 · 30 · 2695-2699
DOI: 10.1109/lcomm.2026.3715317 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Communications Letters, 2026. Verfasst von Yu Jin und 2 weiteren Autor:innen. Die Arbeit nennt 11 Literaturangaben. Gefördert von: Basic and Applied Basic Research Foundation of Guangdong Province, Guangdong Province Graduate Education Innovation Project.
Abstract (Originalsprache)
Spectrum sensing in Non-Orthogonal Multiple Access (NOMA) systems is difficult under power-domain superposition, severe fading, and low SNR. This letter proposes a Hybrid Collaborative Time-Frequency Network (HCTF-Net) for cooperative multi-class sensing from raw secondary user (SU) observations. HCTF-Net integrates hybrid local feature extraction with dual-stream time-frequency discrimination to learn state representations for superposed primary user (PU) signals. Simulation results show that HCTF-Net achieves mean Area Under the Curve (mAUC) values of 0.893 and 0.860 under Rician and Rayleigh fading at −15 dB, respectively, and 0.765 in the dense four PU case, indicating its robustness under low SNR, fading, and dense PU settings.
Förderung
- Basic and Applied Basic Research Foundation of Guangdong Province
- Guangdong Province Graduate Education Innovation Project
Zitieren
APA
Jin, Y., Wang, Y., & Peng, S. (2026). A Hybrid Collaborative Time–Frequency Network for NOMA Spectrum Sensing. IEEE Communications Letters, 30, 2695-2699. https://doi.org/10.1109/lcomm.2026.3715317
Vancouver
Jin Y, Wang Y, Peng S. A Hybrid Collaborative Time–Frequency Network for NOMA Spectrum Sensing. IEEE Communications Letters. 2026;30:2695-2699. doi:10.1109/lcomm.2026.3715317
BibTeX
@article{Jin2026,
author = {Jin, Yu and Wang, Yonghua and Peng, Shiguo},
title = {A Hybrid Collaborative Time–Frequency Network for NOMA Spectrum Sensing},
journal = {IEEE Communications Letters},
year = {2026},
volume = {30},
pages = {2695--2699},
doi = {10.1109/lcomm.2026.3715317}
}
EUVIMED