Design of Fast Radio Burst Signal Recognition System based on Deep Learning

Haoran Yuan; Chungao Shi

doi:10.54691/y8hbe351

Authors

Haoran Yuan
Chungao Shi

DOI:

https://doi.org/10.54691/y8hbe351

Keywords:

Object Detection; Incoherent Dispersion; Fast Radio Bursts; YOLO.

Abstract

In order to identify fast radio burst signals from the original observation data of FAST radio telescopes, this paper designs a fast radio burst signal recognition system based on deep learning object detection algorithm. The system uses the incoherent achromatization algorithm and the YOLO series target recognition algorithm to realize the recognition of fast radio burst signals, and provides users with a friendly graphical system interface. In view of the different performance of users' computers, the system has the function of selecting different algorithm models. Experiments have proved that the system achieves 86% recall and 83% accuracy in the FRB20201124A real-world data test set.

Downloads

Download data is not yet available.

References

Liu Yanling, Chen Maozheng, Yuan Jianping. A review of fast radio burst search methods based on Machine learning [J]. Astronomical Research and Technology, 202, 19(05) : 509-517.

Lorimer D R, Bailes M, Mclaughlin M A, et al. A bright millisecond radio burst of extragalactic origin [J]. Science, 2007, 318(5851): 777-780.

Niu C H, Li D, Luo R, et al. CRAFTS for fast radio bursts extending the dispersion-fluence relation with new FRBs detected by FAST [J]. The Astrophysical Journal Letters, 2021, 909 (1) : L8.

Barsdell B R, Bailes M, Barnes D G, et al. Accelerating incoherent dedispersion [J]. Monthly Notices of the Royal Astronomical Society, 2012, 422(1) : 379-392.

Sclocco A, Leeuwen J V, Bal H E, et al. Real-time dedispersion for fast radio transientsurveys, using auto tuning on many-core accelerators[J]. Astronomy and Computing, 2016, 14 : 1-7.

Ransom S M. New search techniques for binary pulsars[D]. Cambridge : Harvard University, 2001.

MMYOLO Contributors. (2022). MMYOLO: OpenMMLab YOLO series toolbox and benchmark. Retrieved from https://github.com/open-mmlab/mmyolo.

Xu Zhijun, An Tao, Guo Shaoguang et al. A fast radio burst data set for raw data search [J]. Science in China: Physics, Mechanics and Astronomy,2023,53(02):48-59. (in Chinese)

Xu, H., Niu, J.R., Chen, P. et al. A fast radio burst source at a complex magnetized site in a barred galaxy. Nature 609, 685–688 (2022).

Tohti Tinur, Zhang Hailong, Wang Jie, Ye Xinchen. Incoherent dispersion algorithm based on GPU. Astronomical Research and Technology, 21, 18(1) : 60-68.

Connor L, Van Leeuwen J. AJ, 2018, 156: 256.