WF-SwinUnet: A Window Fusion-based RFI Segmentation Model and Its Application in FAST
Open Access
- 7 August 2025
- journal article
- Published by American Astronomical Society in The Astronomical Journal
- Vol. 170 (3) , 151
Abstract
A key challenge in ensuring the quality of high-sensitivity observations from the Chinese Sky Eye FAST is the efficient suppression of radio frequency interference (RFI). However, the traditional thresholding method cannot adapt to the complex baseline noise because it relies on manual parameter adjustment and assumes a constant threshold value, it also faces bottlenecks such as the failure to detect weak interferences and the ambiguous positioning of the boundary. This paper proposes a deep learning framework (WF-SwinUnet) based on the multiwindow fusion Swin Transformer, which achieves end-to-end accurate RFI segmentation through the collaborative design of an adaptive multiscale window mechanism and a bottleneck residual network. Experiments show that compared with the SumThreshold method, this model improves the accuracy, recall and F1 score by 23.4%, 32.1%, and 30.1%, respectively, improves the processing efficiency by 60% (reducing the time from 10,440.95 to 4387.87 ms), and significantly improves the accuracy of boundary positioning, reducing the boundary error (HD95) by 73.6% compared with ArPLS-ST. For weak narrowband RFI (signal strength close to background noise), which is difficult to detect with traditional methods, the model F1 score reaches 0.6808, avoiding signal loss due to threshold overload. In addition, among the six FAST observation targets, the model’s average F1 score reached 0.818, an improvement of 8.3% over the existing best deep learning model (RFI-Net). Ablation experiments verified the synergistic effect of the multiwindow fusion mechanism and the residual module, which together improved the overall performance by 21%.Keywords
Funding Information
- MOST ∣ National Natural Science Foundation of China (12363010)
- Guizhou provincial Basic Projects ((ZK[2023 039))
- Key Technology RD Program (([2023 352))
This publication has 18 references indexed in Scilit:
- Performance of the segment anything model in various RFI/events detection in radio astronomyPublications of the Astronomical Society of Australia, 2025
- Swin Transformer: Hierarchical Vision Transformer using Shifted Windows2021
- The FAST Galactic Plane Pulsar Snapshot survey: I. Project design and pulsar discoveries ⋆Research in Astronomy and Astrophysics, 2021
- Radio frequency interference mitigation based on the asymmetrically reweighted penalized least squares and SumThreshold methodMonthly Notices of the Royal Astronomical Society, 2020
- RFI measurements and mitigation for FASTResearch in Astronomy and Astrophysics, 2020
- Deep residual detection of radio frequency interference for FASTMonthly Notices of the Royal Astronomical Society, 2020
- Selective Kernel Networks2019
- Radio frequency interference mitigation using deep convolutional neural networksAstronomy and Computing, 2017
- Deep Residual Learning for Image Recognition2016
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015