Efficiency of the Least Squares Support Vector Regression in Local ‎Modeling of the Ionosphere Total Electron Content and Comparison ‎with other Models

Mansour Fallah, Tania; Voosoghi, Behzad; Ghaffari-Razin, Seyyed Reza

doi:10.22034/jsst.2024.1454

Efficiency of the Least Squares Support Vector Regression in Local ‎Modeling of the Ionosphere Total Electron Content and Comparison ‎with other Models

Document Type : Original Research Paper

Authors

Tania Mansour Fallah ¹

Behzad Voosoghi ²

Seyyed Reza Ghaffari-Razin ³

¹ M. Sc. Student, Faculty of Geodesy & Geomatics Engineering, K. N. Toosi University of Technology, Tehran, Iran

² Professor, Faculty of Geodesy & Geomatics Engineering, K. N. Toosi University of Technology, Tehran, Iran

³ Assistant Professor, Department of Surveying Engineering, Arak University of Technology, Arak, Iran

10.22034/jsst.2024.1454

Abstract

In this paper, we aim to employ the least squares support vector regression (LS-SVR) for the spatio-temporal modeling of the ionospheric total electron content (TEC). This model utilizes simple linear equations to solve the system of equations, thereby reducing the computational complexity and enhancing both the speed of convergence and the accuracy of the results. We utilized observations from 15 GPS stations in north-western Iran from day 193 to day 228 in 2012. The results of the LS-SVR model were compared with those of support vector regression (SVR), artificial neural networks (ANN), adaptive neuro-fuzzy inference system (ANFIS), Kriging model, global ionospheric maps (GIM), and the International Reference Ionosphere 2016 (IRI2016) as well as TEC values obtained from GPS. The accuracy of all models was evaluated and interpreted at interior and exterior control stations. The analyses indicate that the average root mean square error (RMSE) for the ANN, ANFIS, SVR, LS-SVR, Kriging, GIM, and IRI2016 models at two interior control stations are 3.91, 2.73, 1.27, 1.04, 2.70, 3.02, and 6.93 TECU, respectively. Furthermore, the average relative errors of these models at the same control stations were calculated as 15.98%, 9.39%, 7.85%, 6.09%, 11.60%, 12.54%, and 26.56%, respectively. Analysis of the precise point positioning (PPP) method demonstrated an improvement of 50 mm in the coordinate components using the LS-SVR model. The results of this study demonstrate that the LS-SVR model can serve as a viable alternative to global and empirical models of the ionosphere in the studied area. The LS-SVR model provides a high-precision local ionosphere model.

Keywords

Ionosphere

TEC

GPS

Northwest of Iran

Machine Learning

LS-SVR

Subjects

GPS and navigation GPS)، GLONASS، GALILEO

Article Title Persian

ارزیابی کارایی مدل کمترین مربعات رگرسیون بردار پشتیبان در مدل‌سازی محلی محتوای الکترون کلی یونسفر و مقایسه آن با سایر مدل‌ها

Authors Persian

تانیا منصور فلاح ¹

بهزاد وثوقی ²

سید رضا غفاری رزین ³

¹ دانشجوی کارشناسی ارشد، دانشکده مهندسی نقشه‌برداری، دانشگاه صنعتی خواجه نصیرالدین طوسی، تهران، ایران

² استاد، دانشکده مهندسی نقشه‌برداری، دانشگاه صنعتی خواجه نصیرالدین طوسی، تهران، ایران

³ استادیار، گروه مهندسی عمران و نقشه‌برداری، دانشگاه صنعتی اراک، اراک، ایران

Abstract Persian

در این مقاله هدف استفاده از مدل کمترین مربعات رگرسیون بردار پشتیبان (LS-SVR) جهت مدل‌سازی مکانی-زمانی مقدار محتوای الکترون کلی یونسفر (TEC) است. جهت انجام اینکار، از مشاهدات 15 ایستگاه GPS موجود در منطقه شمالغرب ایران در بازه زمانی روزهای 193 الی 228 از سال 2012 استفاده شده است. مقایسه نتایج مدل جدید با مدل‌های رگرسیون بردار پشتیبان (SVR)، مدل شبکه عصبی مصنوعی (ANN)، مدل استنتاج عصبی-فازی سازگار (ANFIS)، مدل کریجینگ، مدل GIM، مدل تجربی بین‌المللی مرجع یونسفر 2016 (IRI2016) و همچنین مقادیر TEC حاصل از GPS به عنوان مشاهده مرجع انجام می‌گیرد. دقت همه مدل‌ها در ایستگاه‌های کنترل داخلی و خارجی ارزیابی و تفسیر شده است. آنالیزهای انجام گرفته نشان می‌دهد که میانگین RMSE مدل‌های ANN، ANFIS، SVR، LS-SVR، Kriging، GIM و IRI2016 در دو ایستگاه کنترل داخلی به ترتیب برابر با 91/3، 73/2، 27/1، 04/1، 70/2، 02/3 و 93/6 TECU بوده است. تجزیه و تحلیل روش PPP بهبود 50 میلی‌متری در مولفه‌های مختصات با استفاده از مدل LS-SVR را نشان می‌دهد. نتایج این مقاله نشان می‌دهد که مدل LS-SVR را می‌توان به عنوان جایگزینی برای مدل‌های جهانی و تجربی یونسفر در منطقه مورد مطالعه در نظر گرفت. مدل LS-SVR یک مدل یونسفر محلی با دقت بالا محسوب می‌شود.

Keywords Persian

یونسفر

TEC

GPS

شمال غرب ایران

یادگیری ماشین

LS-SVR

[1] I. Sayin, F. Arikan, and O. Arikan, "Regional TEC mapping with random field priors and kriging," Radio Science, vol. 43, no. 5, 2008, Art. no. RS5012, https://doi.org/10.1029/2007RS003786.

[2] M.R. Ghaffari ‎Razin, "Development and analysis of 3D ionosphere modeling using base functions and GPS data over iran," Acta Geodaetica et Geophysica, vol. 51, pp. 95-111, 2016, https://doi.org/10.1007/s40328-015-0113-9.

[3] M. R. Ghaffari Razin and B. Voosoghi, "Regional ionosphere modeling using spherical cap harmonics and empirical orthogonal functions over iran," Acta Geodaetica et Geophysica, vol. 52, pp. 19-33, 2017, https://doi.org/10.1007/s40328-016-0162-8.

[4] Y. Amerian, B. Voosoghi, and M. M. Hossainali, "Regional ionosphere modeling in support of IRI and wavelet using GPS observations," Acta Geophysica, vol. 61, pp. 1246-1261, 2013, https://doi.org/10.2478/s11600-013-0121-5.

[5] H. Etemadfard and M. Mashhadi Hossainali, "Application of slepian theory for improving the accuracy of SH‐based global ionosphere models in the arctic region," Journal of Geophysical Research: Space Physics, vol. 121, no. 3, pp. 2583-2594, 2016, https://doi.org/10.1002/2015JA021811.

[6] M. A. Sharifi and S. Farzaneh, "Regional TEC dynamic modeling based on slepian functions," Advances in Space Research, vol. 56, no. 5, pp. 907-915, 2015, https://doi.org/10.1016/j.asr.2015.05.024.

[7] R. Mautz, J. Ping, K. Heki, B. Schaffrin, C. Shum, and L. Potts, "Efficient spatial and temporal representations of global ionosphere maps over japan using B-spline wavelets," Journal of Geodesy, vol. 78, pp. 662-667, 2005, https://doi.org/10.1007/s00190-004-0432-z.

[8] P. Muhtarov, I. Kutiev, and L. Cander, "Geomagnetically correlated autoregression model for short-term prediction of ionospheric parameters," Inverse Problems, vol. 18, no. 1, 2002, Art. no. 49, https://doi.org/10.1088/0266-5611/18/1/304‎.

[9] P. Nematipour, M. Raoofian-Naeeni, and M. R. Ghaffari Razin, "Regional application of C1 finite element interpolation method in modeling of ionosphere total electron content over europe," Advances in Space Research, vol. 69, no. 3, pp. 1351-1365, 2022, https://doi.org/10.1016/j.asr.2021.11.030‎.

[10] H. Jang and E. Topal, "A review of soft computing technology applications in several mining problems," Applied Soft Computing, vol. 22, pp. 638-651, 2014, https://doi.org/10.1016/j.asoc.2014.05.019.

[11] L. R. Cander, R. Leitinger, and M. Levy, "Ionospheric models including the auroral environment," in Workshop on Space Weather, WPP-155, European Space Agency, Noordwijk, the Netherlands, pp. 135-142, 1999.

[12] J. B. Habarulema, L.A. McKinnell, and B. D. L. Opperman, "Regional GPS TEC modeling; attempted spatial and temporal extrapolation of TEC using neural networks," Journal of Geophysical Research: Space Physics, vol. 116, no. A4, 2011, Art. no. A04314, https://doi.org/10.1029/2010JA016269.

[13] M. R. Ghaffari Razin, B. Voosoghi, and A. Mohammadzadeh, "Efficiency of artificial neural networks in map of total electron content over iran," Acta Geodaetica et Geophysica, vol. 51, pp. 541-555, 2016, http://doi.org/10.1007/s40328-015-0143-3‎.

[14] A. Tebabal, S. Radicella, B. Damtie, Y. Migoya-Orue, M. Nigussie, and B. Nava, "Feed forward neural network based ionospheric model for the east african region," Journal of Atmospheric and Solar-Terrestrial Physics, vol. 191, 2019, Art. no. 105052, https://doi.org/10.1016/j.jastp.2019.05.016.

[15] S. Inyurt and A. Sekertekin, "Modeling and predicting seasonal ionospheric variations in turkey using artificial neural network (ANN)," Astrophysics and Space Science, vol. 364, no. 4, 2019, Art. no. 62, https://doi.org/10.1007/s10509-019-3545-9.

[16] R. Feizi, B. Voosoghi, and M. R. Ghaffari Razin, "Regional modeling of the ionosphere using adaptive neuro-fuzzy inference system in iran," Advances in Space Research, vol. 65, no. 11, pp. 2515-2528, 2020, https://doi.org/10.1016/j.asr.2020.02.027.

[17] C. Cesaroni et al., "Neural network-based model for global total electron content forecasting," Journal of Space Weather and Space Climate, vol. 10, 2020, Art. no. 11, https://doi.org/10.1051/swsc/2020013.

[18] L. Liu, S. Zou, Y. Yao, and Z. Wang, "Forecasting global ionospheric TEC using deep learning approach," Space Weather, vol. 18, no. 11, 2020, Art. no. e2020SW002501, https://doi.org/10.1029/2020SW002501.

[19] G. Xia et al., "Ionospheric TEC forecast model based on support vector machine with GPU acceleration in the china region," Advances in Space Research, vol. 68, no. 3, pp. 1377-1389, 2021, https://doi.org/10.1016/j.asr.2021.03.021.

[20] M. Adolfs and M. M. Hoque, "A neural network-based TEC model capable of reproducing nighttime winter anomaly," Remote Sensing, vol. 13, no. 22, 2021, Art. no. 4559, https://doi.org/10.3390/rs13224559.

[21] A. J. Smola, and B. Schölkopf, "On a kernel-based method for pattern recognition, regression, approximation, and operator inversion," Algorithmica, vol. 22, pp. 211-231, 1998, https://doi.org/10.1007/PL00013831.

[22] M. R. Ghaffari Razin and S. Inyurt, "Spatiotemporal analysis of precipitable water vapor using ANFIS and comparison against voxel-based tomography and radiosonde," GPS Solutions, vol. 26, pp. 1-13, 2022, https://doi.org/10.1007/s10291-021-01184-1.

[23] D. Bilitza, L.A. McKinnell, B. Reinisch, and T. Fuller-Rowell, "The international reference ionosphere today and in the future," Journal of Geodesy, vol. 85, pp. 909-920, 2011.

[24] L. Ciraolo, F. Azpilicueta, C. Brunini, A. Meza, and S. M. Radicella, "Calibration errors on experimental slant total electron content (TEC) determined with GPS," Journal of Geodesy, vol. 81, pp. 111-120, 2007, https://doi.org/10.1007/s00190-006-0093-1.

[25] C. Cortes and V. Vapnik, "Support-vector networks," Machine Learning, vol. 20, pp. 273-297, 1995, https://doi.org/10.1007/BF00994018.

[26] J. Suykens, T. Van Gestel, J. De Brabanter, B. De Moor, and J. Vandewalle, Least Squaress Support Vector Machines, Singapore: World Scientific, 2002, https://doi.org/10.1142/5089.

[27] G. Matheron, The Theory of Regionalized Variables and its Applications, École National supérieure Des Mines, France, 1971.

[28] V. R. Joseph, "Limit kriging, " Technometrics, vol. 48, no. 4, pp. 458-466, 2006, https://doi.org/10.1198/004017006000000011.

[29] S. Erdoğan, "Modelling the spatial distribution of DEM error with geographically weighted regression: an experimental study, " Computers & Geosciences, vol. 36, no. 1, pp. 34-43, 2010, https://doi.org/10.1016/j.cageo.2009.06.005.

[30] J. Li and A. D. Heap, A Review of Spatial Interpolation Methods for Environmental Scientists, Canberra, Australia: Geoscience Australia, 2008.

[31] S. Schaer, Mapping and Predicting the Earth's Ionosphere Using the Global Positioning System, Société helvétique des sciences naturelles, Commission géodésique, Société Helvétique des Sciences Naturelles, Commission Géodésique and Zurich, Switzerland: Institut für Geodäsie und Photogrammetrie, Eidgenössische Technische Hochschule Züric, 1999.

[32] M. S. Nezamzadeh, B. Voosoghi, and S. R. Ghaffari Razin, "Evaluation of the effect of solar and geomagnetic parameters in spatio-temporal modeling of ionosphere's total electron content using machine learning methods," Journal of the Earth and Space physisc, vol. 49, no. 1, pp. 153-169, 2023, https://doi.org/10.22059/jesphys.2023.339441.100‎‎7405‎.