FPN-Transformer with uncertainty head reduces RMSE for cross-height CKM prediction to 5.347 dB zero-shot and 3.518 dB few-shot on a layered aerial benchmark, outperforming 3D-RadioDiff.
Open-set automatic modulation classification using deep metric learning and OpenMax,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Geometry-Aware Cross-Height Channel Knowledge Map Prediction for UAV-Assisted Communications With Uncertainty-Guided 3D Sensing
FPN-Transformer with uncertainty head reduces RMSE for cross-height CKM prediction to 5.347 dB zero-shot and 3.518 dB few-shot on a layered aerial benchmark, outperforming 3D-RadioDiff.