Document Type : Original Article

Authors

1 Faculty of Geoscience Engineering, Arak University of Technology, Arak, Iran

2 Department of Civil engineering, Islamic Azad University of Khoy, Khoy, Iran

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

Abstract

Computerized Ionospheric Tomography (CIT) is a method to reconstruct ionospheric electron density image by computing Total Electron Content (TEC) values from the recorded GPS signals. Due to the poor spatial distribution of GPS stations, limitations of signal viewing angle and discontinuity of observations in time and space domain, CIT are an inverse ill-posed problem. In order to solve these problems, two new methods are developed and compared with the initial method of Residual Minimization Training Neural Network (RMTNN). Modified RMTNN (MRMTNN) and Ionospheric Tomography based on the Neural Network (ITNN) is considered as new methods of CIT. In all two methods, Empirical Orthogonal Functions (EOFs) are used to improve accuracy of vertical domain. Also, Back Propagation (BP) and Particle Swarm Optimization (PSO) algorithms are used to train the neural networks. To apply the methods for constructing a 3D-image of the electron density, 23 GPS measurements of the International GNSS Service (IGS) with different geomagnetic indexes are used. For validate and better assess reliability of the proposed methods, 4 ionosondee stations have been used. Also the results of proposed methods have been compared to that of the NeQuick empirical ionosphere model. Based on the analysis and comparisons, the RMSE of the ITNN model at high geomagnetic activity in DOUR, JULI, PRUH and WARS ionsonde stations are 1.22, 1.46, 1.18 and 1.19 (1011 ele./m3), respectively. The results show that RMSE of the ITNN model is less than other models in both high and low geomagnetic activities and in ionosonde stations.

Keywords

Main Subjects

Article Title [Persian]

Ionospheric electron density reconstruction over central Europe using neural networks: A comparative study

Authors [Persian]

  • Seyyed Reza Ghaffari-Razin 1
  • Reza Davari-Majd 2
  • Behzad Voosoghi 3
  • Navid Hooshangi 1

1 Faculty of Geoscience Engineering, Arak University of Technology, Arak, Iran

2 Department of Civil engineering, Islamic Azad University of Khoy, Khoy, Iran

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

Abstract [Persian]

Computerized Ionospheric Tomography (CIT) is a method to reconstruct ionospheric electron density image by computing Total Electron Content (TEC) values from the recorded GPS signals. Due to the poor spatial distribution of GPS stations, limitations of signal viewing angle and discontinuity of observations in time and space domain, CIT are an inverse ill-posed problem. In order to solve these problems, two new methods are developed and compared with the initial method of Residual Minimization Training Neural Network (RMTNN). Modified RMTNN (MRMTNN) and Ionospheric Tomography based on the Neural Network (ITNN) is considered as new methods of CIT. In all two methods, Empirical Orthogonal Functions (EOFs) are used to improve accuracy of vertical domain. Also, Back Propagation (BP) and Particle Swarm Optimization (PSO) algorithms are used to train the neural networks. To apply the methods for constructing a 3D-image of the electron density, 23 GPS measurements of the International GNSS Service (IGS) with different geomagnetic indexes are used. For validate and better assess reliability of the proposed methods, 4 ionosondee stations have been used. Also the results of proposed methods have been compared to that of the NeQuick empirical ionosphere model. Based on the analysis and comparisons, the RMSE of the ITNN model at high geomagnetic activity in DOUR, JULI, PRUH and WARS ionsonde stations are 1.22, 1.46, 1.18 and 1.19 (1011 ele./m3), respectively. The results show that RMSE of the ITNN model is less than other models in both high and low geomagnetic activities and in ionosonde stations.

Keywords [Persian]

  • Total Electron Content
  • Tomography
  • Residual Minimization Training Neural Network
  • Ionospheric Tomography based on the Neural Network
  • GPS
[1] J. R., Austen, S.J., Franke, C. H., Liu, "Ionospheric imaging using computerized tomography". Radio Sci., vol. 23, pp. 299–307, 1988.
[2] V.E., Kunitsyn, I.A., Nesterov, A. M., Padokhin, U. S., Tumanova, "Ionospheric radio tomography based on the GPS/GLONASS navigation systems". J Commun Technol Electron, vol. 56, pp. 1269–1281, 2011.
[3] D., Pokhotelov, P., Jayachandran, C. N., Mitchell, J. W., MacDougall, M. H., Denton, "GPS tomography in the polar cap: comparison with ionosondees and in situ spacecraft data". GPS Solut, vol. 15, pp. 79–87, 2011.
[4] D. B., Wen, Y., Wang, R., Norman, "A new two-step algorithm for ionospheric tomography solution,” GPS Solut., vol. 16, pp. 89-94, 2012.
[5] M.M.J.L., Van de Kamp, "Medium-scale 4-D ionospheric tomography using a dense GPS network". Ann Geophys., vol. 31, pp. 75-89, 2013.
[6] M.R., Ghaffari Razin, "Development and analysis of 3D ionosphere modeling using base functions and GPS data over Iran". Acta Geod Geophys, vol.  51, pp. 95-111, 2015.
[7] M.R., Ghaffari Razin, B., Voosoghi, "Regional ionosphere modeling using spherical cap harmonics and empirical orthogonal functions over Iran". Acta Geod. Geophys., vol. 52, pp. 19-33, 2016.
[8] A., Yilmaz, K.E., Akdogan, M., Gurun, "Regional TEC mapping using neural networks". Radio Sci, vol. 44, pp. 1-16 , 2009.
[9] X.F., Ma, T., Maruyama, G., Ma, T., Takeda, "Three dimensional ionospheric tomography using observation data of GPS ground receivers and ionosondee by neural network". J. Geophys. Res., vol. 110, pp. 1-12, 2005.
[10] S., Hirooka, K., Hattori, T., Takeda, "Numerical validations of neural-network-based ionospheric tomography for disturbed ionospheric conditions and sparse data". Radio Sci., vol. 46, 2011.
[11] M. R., Ghaffari Razin,  B., Voosoghi, "Regional application of multi-layer artificial neural networks in 3D ionosphere tomography". Adv Space Res, vol.  58 pp. 339–348, 2016.
[14] L., Ciraolo, F., Azpilicueta, C., Brunini, A., Meza, S. M., Radicella, "Calibration errors on experimental Slant Total Electron Content (TEC) determined with GPS". Journal of Geodesy, vol. 81, pp:111–120, 2007.
[15] A., Liaqat, M., Fukuhara, T., Takeda, "optimal estimation of parameter of dynamical system by neural network collocation method". Comput. Phys. Commun, vol. 150, pp. 215–234, 2003.
[16] G., Seeber, "Satellite Geodesy, Foundations, Methods and Application", Walter de Gruyter, Berlin and New York, 531.
[17] A., Quarteroni, R., Sacco, F., Saleri, "Numerical Mathematics, 37, Texts in Applied Mathematics", Springer Berlin Heidelberg, Heidelberg, Germany, 2007.
[18] A., Alexandridis, A., Zapranis, "Wavelet neural networks: A practical guide". Neural Networks, vol.  42, pp, 1–27, 2013.
[19] Y., Becerikli, Y., Oysal, A. F., Konar, "On a dynamic wavelet network and its modeling application". In Lecture Notes in Computer Science, vol, 2714, pp. 710–718, 2003.
[20] S. A., Billings, H. L., Wei, "A new class of wavelet networks for nonlinear system identification". IEEE Transactions on Neural Networks, vol. 16, pp. 862–874, 2005.
[21] Y., Oussar, G., Dreyfus, "Initialization by selection for wavelet network training". Neurocomputing, vol. 34, pp. 131–143, 2000.
[22] L., Xu, D. W. C., Ho, "A basis selection algorithm for wavelet neural networks". Neurocomputing, vol. 48, pp. 681–689, 2002.
[23] Q., Zhang, A., Benveniste, "Wavelet Networks". IEEE Trans. Neural Networks, vol. 3, pp.889–898, 1992.
[24] M. R., Ghaffari Razin,  B., Voosoghi, " Wavelet neural networks using particle swarm optimization training in modeling regional ionospheric total electron content". Journal of Atmospheric and Solar-Terrestrial Physics, vol. 149, pp. 21-30, 2016.
[25] M. R., Ghaffari Razin,  B., Voosoghi, " Modeling of ionosphere time series using wavelet neural networks (case study: NW of Iran)". Advances in Space Research, vol. 58, pp. 74-83, 2016.