A lightweight neural precoder trained with teacher-student distillation and meta-learning generalizes across simulated city sites and, after fine-tuning, reportedly outperforms WMMSE with lower computational cost.
Unfolding WMMSE using graph neural networks for efficient power allocation,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
eess.SP 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
A Low-Complexity Plug-and-Play Deep Learning Model for Massive MIMO Precoding Across Sites
A lightweight neural precoder trained with teacher-student distillation and meta-learning generalizes across simulated city sites and, after fine-tuning, reportedly outperforms WMMSE with lower computational cost.