عنوان مقاله [English]
Ionosphere is a region of ionized plasma that extends at an altitude of 80 to 1,200 km above the earth's surface. The ionosphere consists of free electrons and ions formed during the ionization process. Total electron content (TEC) in the ionosphere is reported in TECU units. Each TECU is equivalent to 1016 electron units per square meter. Ionosphere is highly sensitive to any atmospheric turbulence, and thus is considered to be an atmospheric event sensor. The present study seeks to investigate the effect of space and temperature on the amount of total ionospheric electron content in order to accurately estimate TEC value. To reach this aim, variations in latitude and longitude are decomposed for a given period of time using the process of transforming wavelet to frequency component and modeled using a variety of artificial neural networks.
Materials and Methods
Here, after separating the location and temperature parameters in each region, ionospheric electron density is estimated for each spatial and temperature parameter separately and also as a combination using the capabilities of artificial neural networks and wavelet transform. TEC value for each location and temperature parameter is extracted from the ionospheric maps and then used as input data in the suggested method. These maps show ionospheric electron content. The standard format of ionospheric maps, which contains TEC values is called IONEX. These files are received from the website of Iranian National Mapping Agency.
Results and discussion
In general, IONEX is divided into three different parts: description, TEC maps, and standard deviations of maps. TEC values are presented in a regular network. Each IONEX file includes 25 maps, the last of which is the first map of the next day. As mentioned before, TEC value gives us a better understanding of ionospheric behavior. Availability of enough data and time coverage are two important factors in understanding a phenomenon and proper evaluation of its behavior.
As results of artificial neural networks indicate, MLP generally has lower RMSE values. Therefore, it gives a more accurate estimation of TEC, compared to other artificial neural networks. Also compared to artificial neural networks, a combination of artificial neural networks and wavelet shows better results. The best condition of all three methods shows that compared to other methods, temperature variations give us a better estimation of TEC in ionosphere.
6. A. R. Amiri-Simkooei et al., 2012, Harmonic analysis of total electron contents time series: methodology and results. GPS Solution, pages 77-88.
7. Amerian, Y., MashhadiHossainali, M., and Voosoghi, B., Ghafari, M.R “Tomographic Reconstruction of the Ionospheric Electron Density in term of Wavelets.” Iranian Aerospace Society7 (1): 19-29.
8. Amerian Yazdan, 2010,behzad voosoghi and masoud m. hossainali. Regional Ionosphere Modeling in Support of IRI and Wavelet Using GPS Observations. Tehran, Iran: K.N. Toosi University of Technology.
9. Anghel, A. F., 2009, studies of the thermosphere, ionosphere, and plasma sphere using wavelet analysis, neural networks, and kalman filters, Electrical and Computer Engineering. Colorado, University of Colorado at Boulder, Doctor of Philosophy.
10. Cander. R, 2007 ,Artificial neural network applications in ionospheric studies, Annali di Geofisica, Vol.5-6, 1998Leandro, R. F. A New Technique to TEC Regional Modeling using a Neural Network. Fredericton, Canada, Department of Geodesy and Geomatics Engineering,
11. El-naggar .A. 2011, Enhancing the accuracy of GPS point positioning by converting the single frequency data to dual frequency data, Alexandria Engineering Journal 50, 237–243.
12. GhaffariRazin, M. R., Voosoghi, B., Mohammadzadeh, A”Efficiency of artificial neural networks in map of total electron content over Iran.”Acta GeodGeophys.
13. Ghazouly, A. A. 2013, Multi-Resolution Spectral Techniques for Static DGPS Error Analysis and Mitigation, geomatics engineering, calgary, alberta. Doctor of philosophy.
14. Kurihara, J., Ogawa, Y., Oyama, S., Nozawa, S., Tsutsumi, M., Hall, C.M., Tomikawa, Y.,Fujii, R., 2010. Links between a stratospheric sudden warming and thermal structures and dynamics in the high-latitude mesosphere, lower thermosphere, and ionosphere. Geophys. Res. Lett. 37, L13806.
15. Leandro, R. F. 2007, A New Technique to TEC Regional Modeling using a Neural Network. Fredericton, Canada, Department of Geodesy and Geomatics Engineering.
16. McKinnell, L, A neural network based ionospheric model for the Bottomside electron density profile over Grahamstown South Africa, 2002, Ph.D. Thesis, Rhodes Université.
17. Pancheva, D., Mukhtarov, P., 2012a. Global response of the ionosphere to atmospheric tides forced from below: Recent progress based on satellite measurements. Space Sci. Rev. 168 (1–4), 175–209.
18. Plamen Mukhtarov, Dora Pancheva, 2015, Winter-time dependence of the global TEC on the stratospheric temperature and solar radiation,2015, Journal of Atmospheric and Solar-Terrestrial Physics 136, 134–149.
19. Potts L, Schmidt M, Shum CK, Ge S, 2003, Wavelet based regional multi-resolution TEC model. (abstract),Europe and Geophysical Society–American Geophysical Union–European Union of Geosciences Joint Assembly, Nice, April.
20. R. Leandro, M. Santos, 2004. Regional Computation of TEC using a Neural Network Model. University of New Brunswick, Department of Geodesy and Geomatics Engineering, Fredericton, N.B., E3B 5A3, Canada
21. Salamonowicz PH, 2001, A wavelet-based gravity model with an application to the evaluation of Stokes’ integral. In: Sideris MG(ed) Gravity, geoid and geodynamics 2000. Springer, Berlin Heidelberg NewYork, pp 85–90
22. Schaffrin B, Mautz R, Shum CK, Tseng H, 2003, Towards a spherical pseudo-wavelet basis for geodetic applications. Comput Aided Civ Inf 18(5):369–378.
23. Seeber, G. 2003, Satellite Geodesy, Berlin, Die Deutsche Bibliothek.
24. Tulunay, E., Senalp, E. T., Radicella, S. M. and Tulanay, Y., 2006, Forecasting total electron content maps by neural network technique, Radio Sci. 41, doi: 10.1029/2005RS003285.
25. Xenos, T. D., Kouris, S. S. and Casimiro, A., 2003, Time-dependent prediction degradation assessment of neural-networks- based TEC forecasting models, Nonlinear Proc. Geophys., 10, 585-587.