.png){kind=icon}
.png){kind=icon}
Return
Neural network-based framework for signal separation in spatio-temporal gravity data
.png){kind=icon}
OA
AI
DOI：10.1016/j.cageo.2025.106057.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Supervised training of a neural network-based algorithm separating spatio-temporal gravity signals from their sum. • Flexible implementation of a multi-channel U-Net similar to Kadandale et al. (2020), adapted for the context of signal separation in GRACE-type level 2b satellite gravity data. • Implementation of sampling methods for building 2-D training samples from spatio-temporal geodetic data. • Closed-loop simulation framework for validating the results using synthetic data.
Keywords:
Geospatial AI
Signal separation
Satellite gravity data
GRACE
Neural networks
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
C
IF:
4.4
Papers:
5.0K
Citations:
1.5W
Organization
No organization information available