Diffusion-based denoising of stepwise signals
Abstract
This code can be used for training, validating, and testing diffusion-based models for denoising stepwise signals. The model reconstructs noisy-free signals from noisy stepwise observations and supports signals with different numbers of discrete states.The code includes pretrained models for direct inference as well as scripts for training the model from scratch, and can be applied to both simulated and experimental stepwise signals.
Fields of Study & Subjects
Image and Signal Denoising Methods
Control Systems and Identification
Advanced Adaptive Filtering Techniques
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 29 January 2026
- Language
- EN