A hybrid CNN-Transformer with Dynamic Variance Gate and physics-aware augmentation achieves 97.6% accuracy in NLoS and 98.8% in unseen environments for cross-domain WiFi fall detection without fine-tuning.
Vision transformers for human activity recognition using wifi channel state information,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.SP 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Robust Cross-Domain WiFi Fall Detection via Physics-Driven Attention-Enhanced Transformers
A hybrid CNN-Transformer with Dynamic Variance Gate and physics-aware augmentation achieves 97.6% accuracy in NLoS and 98.8% in unseen environments for cross-domain WiFi fall detection without fine-tuning.