Microwavenet: Lightweight CBAM-Augmented Wavelet-Attentive Networks for Robust EEG Denoising

Artifact contamination in EEG signals poses significant challenges for clinical diagnosis, computational analysis, and in brain-computer interfaces. To address this, we introduce MicroWaveNet, an ultra-lightweight deep learning model ($\sim 60 \mathrm{k}$parameters) for EEG artifact removal. It employs a multi-branch Morlet wavelet transform to separately analyze EEG frequency components, effectiv…