Using a 1D-ResNet with squeeze-and-excitation blocks on 2531-dimensional features from 5G NR SRS, the system achieves 3.92% equal error rate for distinguishing legitimate from spoofed probes on real over-the-air data with chronological splitting.
Enhancing security in 5G NR with channel-robust RF fingerprinting leveraging SRS for cross-domain stability,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.SP 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Deep Learning-Based Physical Layer Authentication Using 5G NR Sounding Reference Signals: A Temporal Generalization Study on Real Testbed Data
Using a 1D-ResNet with squeeze-and-excitation blocks on 2531-dimensional features from 5G NR SRS, the system achieves 3.92% equal error rate for distinguishing legitimate from spoofed probes on real over-the-air data with chronological splitting.