|
|
|
Ionospheric vertical total electron content prediction model in low-latitude regions based on long short-term memory neural network |
| Tong-Bao Zhang(张同宝)1,2, Hui-Jian Liang(梁慧剑)2,3, Shi-Guang Wang(王时光)1,2,†, and Chen-Guang Ouyang(欧阳晨光)1 |
1 Department of Precision Instrument, Tsinghua University, Beijing 100084, China;
2 State Key Laboratory of Precision Measurement Technology and Instrument, Tsinghua University, Beijing 100084, China;
3 Department of Electronic Engineer, Tsinghua University, Beijing 100084, China |
|
|
|
|
Abstract Ionosphere delay is one of the main sources of noise affecting global navigation satellite systems, operation of radio detection and ranging systems and very-long-baseline-interferometry. One of the most important and common methods to reduce this phase delay is to establish accurate nowcasting and forecasting ionospheric total electron content models. For forecasting models, compared to mid-to-high latitudes, at low latitudes, an active ionosphere leads to extreme differences between long-term prediction models and the actual state of the ionosphere. To solve the problem of low accuracy for long-term prediction models at low latitudes, this article provides a low-latitude, long-term ionospheric prediction model based on a multi-input-multi-output, long-short-term memory neural network. To verify the feasibility of the model, we first made predictions of the vertical total electron content data 24 and 48 hours in advance for each day of July 2020 and then compared both the predictions corresponding to a given day, for all days. Furthermore, in the model modification part, we selected historical data from June 2020 for the validation set, determined a large offset from the results that were predicted to be active, and used the ratio of the mean absolute error of the detected results to that of the predicted results as a correction coefficient to modify our multi-input-multi-output long short-term memory model. The average root mean square error of the 24-hour-advance predictions of our modified model was 4.4 TECU, which was lower and better than 5.1 TECU of the multi-input-multi-output, long short-term memory model and 5.9 TECU of the IRI-2016 model.
|
Received: 16 September 2021
Revised: 13 December 2021
Accepted manuscript online: 18 December 2021
|
|
PACS:
|
07.05.Mh
|
(Neural networks, fuzzy logic, artificial intelligence)
|
| |
84.35.+i
|
(Neural networks)
|
| |
94.20.dt
|
(Equatorial ionosphere)
|
| |
96.12.ji
|
(Ionospheres)
|
|
| Fund: We would like to thank Data Center for Meridian Space Weather Monitoring Project (NSSC, CAS) for providing detected ionospheric vTEC data, and to thank Space Physics Data Facility of Goddard Space Flight Center (NASA) for providing calculated Dst data and calculated ionospheric vTEC data. We would like to thank Zhengbo Wang, Jianwei Zhang, and Yuhang Li for feedback on our manuscript. Project supported by the National Key Research and Development Program of China (Grant No. 2016YFA0302101) and the Initiative Program of State Key Laboratory of Precision Measurement Technology and Instrument. |
Corresponding Authors:
Shi-Guang Wang
E-mail: wangsg@tsinghua.edu.cn
|
Cite this article:
Tong-Bao Zhang(张同宝), Hui-Jian Liang(梁慧剑),Shi-Guang Wang(王时光), and Chen-Guang Ouyang(欧阳晨光) Ionospheric vertical total electron content prediction model in low-latitude regions based on long short-term memory neural network 2022 Chin. Phys. B 31 080701
|
[1] Hoque M M and Jakowski N 2006 J. Geod. 81 259
[2] Anna K G, Pawel W and Andrzej B 2017 Remote Sens. 9 1221
[3] Hadas T, Krypiak G A, Hernández P M, et al. 2017 J. Geop. 122 9420
[4] Li M, Yuan Y, Wang N, et al. 2018 J. Geod. 92 889
[5] Fujieda M, Piester D, Gotoh T, Becker J, Aida M and Bauch A 2014 Metrologia 51 253
[6] Cyril G, Donald D B and Dustin M S 2015 Planetary and Space Science 117 421
[7] Laurence G A, Karim E M and George S 2000 Geophysical Research Letters 27 1451
[8] Ji Y, Zhang Y, Dong Z, Zhang Q, Li D and Yao B 2020 IEEE Transactions on Geoscience and Remote Sensing 58 3941
[9] Ilyushin Y A 2009 Radiophys Quant. Electron. 52 332
[10] Alizadeh M M, Wijaya D D, Hobiger T, Weber R and Schuh H 2013 Atmospheric effects in space geodesy Springer, Berlin, Heidelberg. pp. 35-71
[11] Su K, Jin S, Jiang J, et al. 2021 GPS Solut. 25 68
[12] http://arxiv.org/abs/1810.13273
[13] Sun W, Xu L, Huang X, et al. 2017 2017 IEEE Visual Communications and Image Processing 10-13 December 2017, St. Petersburg, FL, USA, 17582618
[14] Gruet M A, Chandorkar M, Sicard A, et al. 2018 Space Weather 16 1882
[15] Srivani I, Prasad G S V, Ratnam D V 2019 IEEE Geoscience and Remote Sensing Letters 16 1180
[16] https://avesis.kocaeli.edu.tr/yayin/78af7cc4-a6af-4a3c-a18d-cafba49674dd/ionospheric-tec-prediction-performance-of-arima-and-lstm-methods-in-different-space-weather-conditions
[17] Tang R, Zeng F, Chen Z, et al. 2020 Atmosphere 11 316
[18] Ruwali A, Kumar A J S, Prakash K B, et al. 2020 IEEE Geoscience and Remote Sensing Letters 18 1004
[19] Liu L, Zou S, Yao Y, et al. 2020 Space Weather 18 e2020SW002501
[20] Rao T V, Sridhar M, Ratnam D V, et al. 2021 IEEE Geoscience and Remote Sensing Letters
[21] Wen Z, Li S, Li L, et al. 2021 Astrophysics and Space Science 366 1
[22] Xiong P, Zhai D, Long C, et al. 2021 Space Weather 19 e2020SW002706
[23] Salar F W, Chowdary J D, Reddy C R, et al. 2021 Materials
[24] Moon S, Kim Y H, Kim J H, et al. 2020 Journal of the Korean Physical Society 77 1265
[25] Kim J H, Kwak Y S, Kim Y H, et al. 2020 Space Weather 18 e2020SW002590
[26] Kim J, Kwak Y S, Kim Y H, et al. 2021 ESSOAr
[27] Zivot E and Wang J 2003 Rolling Analysis of Time Series (New York:Springer) pp. 299-346
[28] Ivan K, Yuichi O, Dora P and Rod H 2012 J. Geophys. Res. 117 A08330 |
| No Suggested Reading articles found! |
|
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
Altmetric
|
|
blogs
Facebook pages
Wikipedia page
Google+ users
|
Online attention
Altmetric calculates a score based on the online attention an article receives. Each coloured thread in the circle represents a different type of online attention. The number in the centre is the Altmetric score. Social media and mainstream news media are the main sources that calculate the score. Reference managers such as Mendeley are also tracked but do not contribute to the score. Older articles often score higher because they have had more time to get noticed. To account for this, Altmetric has included the context data for other articles of a similar age.
View more on Altmetrics
|
|
|
