Deterministic denoising of long-duration gravitational-wave signal candidates with α-(de)blending
Abstract
We present a deterministic machine-learning method for denoising candidate signals from all-sky searches for continuous gravitational waves from rotating neutron stars. Using the time-domain F-statistic search pipeline, we post-process the resulting F(f,ḟ) candidate patterns with iterative α-(de)blending, a deterministic diffusion-type generative model implemented as a U-Net, well suited to the non-Gaussian, correlated noise of the F-statistic output. Two models, trained on data based on simulated 6- and 12-day time-domain segments with software-injected signals similar to the LVK hardware injections, are tested by comparing deblended images to the library of expected signal patterns via the Structural Similarity Index Measure. In general the method recovers the correct sky-position-dependent pattern for signal-to-noise ratios ρ≳ 4, moderately below typical all-sky detection thresholds, with recovery depending strongly on pattern morphology, demonstrating that deterministic diffusion-based denoising may serve as a consistency/veto tool ahead of the semi-coherent coincidence stage. We also discuss limitations and possible improvements of the method.
Related
- Survey of Novel Deep Learning Architectures for Denoising Gravitational-wave Signals
- Fast Bayesian Inference for Long-Duration Gravitational-Wave Signals in 3G detectors
- Prospects for Simultaneously Detecting Long and Short Duration Gravitational Waves from Pulsar Glitches with Current and Future Detectors
- Probabilistic characterization of blending with LSST and application to cluster lensing cosmology
- Denoising Interferometric Observations Using Visibility-Informed Neural Networks
BibTeX
@misc{figura2026deterministic,
title = {{Deterministic denoising of long-duration gravitational-wave signal candidates with $α$-(de)blending}},
author = {Przemysław Figura and Michał Bejger},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.15308},
eprint = {2609.15308},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.15308},
}
RIS
TY - UNPB
TI - Deterministic denoising of long-duration gravitational-wave signal candidates with α-(de)blending
AU - Przemysław Figura
AU - Michał Bejger
PY - 2026
DA - 2026/09/14
PB - arXiv
DO - 10.48550/arxiv.2609.15308
UR - https://arxiv.org/abs/2609.15308
AB - We present a deterministic machine-learning method for denoising candidate signals from all-sky searches for continuous gravitational waves from rotating neutron stars. Using the time-domain F-statistic search pipeline, we post-process the resulting F(f,ḟ) candidate patterns with iterative α-(de)blending, a deterministic diffusion-type generative model implemented as a U-Net, well suited to the non-Gaussian, correlated noise of the F-statistic output. Two models, trained on data based on simulated 6- and 12-day time-domain segments with software-injected signals similar to the LVK hardware injections, are tested by comparing deblended images to the library of expected signal patterns via the Structural Similarity Index Measure. In general the method recovers the correct sky-position-dependent pattern for signal-to-noise ratios ρ≳ 4, moderately below typical all-sky detection thresholds, with recovery depending strongly on pattern morphology, demonstrating that deterministic diffusion-based denoising may serve as a consistency/veto tool ahead of the semi-coherent coincidence stage. We also discuss limitations and possible improvements of the method.
ER -