پیش‌بینی محتوای کلی الکترون قائم یون‌سپهری با شبکۀ عصبی برای یک موقعیت ‌خاص و مقایسه با مدل مرجع یون‌سپهری بین‌المللی

نویسندگان

1 کارشناسی ارشد ژئودزی-هیدروگرافی، گروه مهندسی نقشه‌برداری، پردیس دانشکده‌های فنی دانشگاه تهران، ایران

2 1. دانشیار، تهران، گروه مهندسی نقشه‌برداری، پردیس دانشکده‌های فنی دانشگاه ایران 2. پژوهشکدۀ مهندسی فناوری‌های اطلاعات مکانی، پردیس دانشکده‌های فنی، دانشگاه تهران

3 استادیار، گروه مهندسی نقشه‌برداری، پردیس دانشکده‌های فنی دانشگاه تهران، ایران

4 دانشجوی دکتری ژئودزی، گروه مهندسی نقشه‌برداری، پردیس دانشکده‌های فنی دانشگاه تهران، ایران

چکیده

با ظهور انواع ماهواره‌ها در چند دهۀ اخیر، مطالعۀ لایۀ یون‌سپهر به یکی از مهم‌ترین موضوع‌ها در علوم مختلف تبدیل شده است؛ چراکه امواج ارسالی از این ماهواره‌ها به سمت زمین ناگزیر از یون‌سپهر عبور می‌کنند. خطای یون‌سپهری یکی از مهم‌ترین عوامل ایجاد خطا در اندازه‌گیری‌های تعیین موقعیت و ناوبری با  GPSمحسوب می‌شود، به‌طوری‌که برای ناوبری دقیق، به داشتن تأخیر یون‌سپهری نیاز است. گیرنده‌های دوفرکانسه قادرند بخش عمده‌ای از این تأخیر را محاسبه کنند ولی در مواردی که فقط از اطلاعات یک فرکانس استفاده می‌شود یا گیرندۀ دوفرکانسه در دسترس نیست، لازم است به طریقی مدل و اثر این خطا را که در مواقع حداکثر فعالیت خورشیدی به چند ppm نیز می‌رسد، تا حد امکان کاهش دهیم. پیچیدگی تغییرات در لایۀ یون‌سپهری موجب عدم حذف کامل اثر یون‌سپهری شده است. در گیرنده‌های تک‌فرکانسه به‌منظور کاهش خطای یون‌سپهری می‌توان از مدل یون‌سپهری (کلوبوچار) موجود در پیغام ناوبری ارسال‌شده از ماهواره استفاده کرد. در این تحقیق کاربرد شبکۀ عصبی در مدل‌سازی و پیش‌بینی محتوای کلی الکترون قائم در بالای منطقه‌ای واقع در استان سیستان و بلوچستان (ایرانشهر) کشور ایران برای سال 2006 که فعالیت خورشیدی در سطح پایینی بوده، بررسی شده است. شبکۀ عصبی پسخور با یک لایۀ پنهان و الگوریتم انتشار روبه‌عقب طراحی شده است و پارامترهای فضای ورودی شبکۀ عصبی تغییرات روزانه، تغییرات فصلی، فعالیت‌های خورشیدی و ژئومغناطیسی می‌باشند. این مدل با محتوای کلی الکترون قائم GPS و مدل IRI2007 در چهار زمان انقلابین و اعتدالین مقایسه شده و در تمامی این زمان‌ها مدل شبکۀ عصبی دقیق‌تر از مدل IRI2007 برای کشور ایران عمل کرده است.

کلیدواژه‌ها

موضوعات


عنوان مقاله [English]

Vertical Total Electron Content (VTEC) prediction with neural network for single station in Iran and comparison with IRI

نویسندگان [English]

  • Farideh Sabzehee 1
  • Mohammad ali Sharifi 2
  • Mehdi Akhoond zadeh 3
  • Saeed Farzaneh 4
1 Department of Surveying and Geomatics Engineering, College of Engineering, University of Tehran.
2 1) Department of Surveying and Geomatics Engineering, College of Engineering, University of Tehran. 2) Research Institute of Geoinformation Technology (RIGT), College of Engineering, University of Tehran, Iran
3 Department of Surveying and Geomatics Engineering, College of Engineering, University of Tehran.
4 Department of Surveying and Geomatics Engineering, College of Engineering, University of Tehran.
چکیده [English]

The ionosphere as the upper part of Earth’s atmosphere consists of electrons and atoms affecting the signal propagation in the radio frequency domain. Nowadays, Global Navigation Satellite Systems (GNSS), like GPS, are widely used for various applications. The majority of navigation satellite receivers operate on a single frequency and experience an error due to the ionospheric delay. They compensate for the ionospheric delay using an ionospheric model which typically only corrects for 50% of the delay. An alternative approach is to map the ionosphere with a network of real-time measurements. Global Positioning System (GPS) networks prepare chance to study the dynamics and continuous variations in the ionosphere by complementary ionospheric measurements, which are usually obtained by different techniques such as ionosondes, incoherent scatter radars and satellites. The ionospheric delay is characterized by the Total Electron Content (TEC) along the signal path from the satellite to the receiver. Elimination (or reduction) of the ionosphere effects is possible using dual-frequency receivers by a very useful combination of dual-frequency data known as geometry free combination (L4) as follows:
                                                                                     
where the L4 signature is derived from L1(1.57542 GHz) and L2(1.22760 GHz) phase observables. Single-frequency users, however cannot take advantage of this combination. So, they have to use a proper ionospheric model to correct the ionospheric delay. Ionosondes (up to an altitude of 1000 km) can determine TEC while GPS measurements give completely information about the topside ionosphere. In this paper, the suitability of Neural Networks (NNs) in order to predict the Total Electron Content (TEC) obtained from Iranian Permanent GPS Network (IPGN) during the low-solar-activity period 2006 has been investigated. TEC has many non-linear variations while the neural network has a significant ability to model and approximate it (Williscroft and Poole, 1996; Hernandez-Pajares et al., 1997; Xenos et al., 2003; Sarma and Mahdu, 2005; Leandro and Santos, 2007). The input space included the day number (DN,seasonal variation), hour (HR,diurnal variation), sunspot number (SSN,measure of the solar activity) and magnetic index (measure of magnetic activity).
To make the data continuous, the first two parameters were each split into sine and cosine components, two cyclic as follows:
 
                                                         
where DNS, DNC, HRS and HRC are the sine and cosine components of DN and HR, respectively.
 
In this paper, the TEC values have been estimated using the PPP (Precise Point Positioning) module of the Bernese over Iranshahr (27˚N, 60˚E).
Optimum situation of the neural network include of single hidden layer and eight neurons of inputs layer and fifty neurons of hidden layer and one neuron of output layer. To this end, the single hidden layer feed-forward network with a back propagation algorithm has been designed.
An analysis was done by comparing predicted NN TEC(TEC values predicted by the NN model) with TEC values from the IRI2007 version of the International Reference Ionosphere, validating GPS TEC(TEC values calculated from the GPS measurements) with the maximum electron density obtained from ionosonde and calculating the performance of the NN model during equinoxes and solstices.
The results show high correlation with GPS TEC and NN TEC. Their Root-Mean-Square Error(RMSE) and coefficient of determination (R2) are 1.5273 TECU and 0.9334 respectively.
RMSE is defined as:
 
 
 
 
where N is the number of data points.
 
In table 3, the absolute error (Eabs) is defined as the magnitude of the difference between the NN predicted TEC and the GPS TEC, while the relative error(Erel) is the ratio of the absolute error to GPS TEC and can be represented as a percentage (Habarulema et al, 2007).These errors were calculated as follows:
 
 
 
 is the absolute error and  is the relative error respectively. The difference (100- % gives the relative correction, which indicates the approximate TEC prediction accuracy for the NN model (Leandro and Santos, 2007). An average error of ~11.41% means that the NN can predict about 88.58% of the GPS TEC on average. Results show that the neural network works better rather than the IRI model for IRAN. 

کلیدواژه‌ها [English]

  • Vertical total electron content
  • Neural Network
  • International Reference Ionosphere (IRI)
  • Global positioning System (GPS)
Bilitza, D.,2001, International Reference Ionosphere 2000, Radio Sci., 36(2), 261–275, doi:10.1029/2000RS002432.
Fausett, L.,1994, Fundamentals of Neural Networks: Architectures, Algorithms, and Applications. Prentice Hall, New Jersey.
Habarulema, J.B., McKinnell, L.A., and Cilliers, P.J., 2007, Prediction of global positioning system total electron content using neural networks over South Africa, J. Atmos Solar Terr. Phys., 69, 1842–1850.
 
Hargreaves, J. K., 1992, The Solar-Terrestrial Environment: an introduction to geospace - the science of the terrestrial upper atmosphere, ionosphere and magnetosphere, Cambridge University Press,Cambridge.
 
Haykin, S.,1994, Neural Networks, A Comprehensive Foundation, Macmillan College Publishing Company.
 
Hofmann-Wellenhof, B., Lichtenegger, H., Collins, J., 1992, Global Positioning System theory and Practice, Springer-Verlag, Wien, New York.
Kelley, M. C.,2009, The Earth's Ionosphere: Plasma Physics and Electrodynamics, Intern. Geophys.Series, vol 43. San Diego: Academic Press.
 Leandro R.F. and Santos M., 2004, Regional Computation of TEC using a Neural Network Model, Poster presented at the Joint Assembly of CGU, AGU, SEGand EEGS, Montreal.
 
Leandro, R.F., Santos, M.C.,2007, A neural network approach for regional vertical total electron content modelling, Studia Geophysica et Geodaetica, 51, 279–292.
 
McKinnell, L.A.,2002, A Neural Network based Ionospheric model for the bottomside electron density profile over Grahamstown, South Africa, Ph.D. thesis of Rhodes University, Grahamstown, South Africa.
 
McKinnell L.A., Opperman B. and Cilliers P.J., 2007, GPS TEC and Ionosonde TEC over Grahamstown, South Africa: First Comparisons, Advances in Space Research, 39, pp. 816–820.
 
Opperman, B.,2007, Reconstructing Ionospheric TEC over South Africa using signals from a Regional GPS network, PhD thesis Rhodes University, Grahamstown, South Africa.
 
 
Opperman, B.D.L., Cilliers, P.J., McKinnell, L.A., and Haggard, R.,2007, Development of a regional GPS-based ionospheric TEC model for South Africa, Adv. Space Res., 39, 808–815.
 
Poole A.W. and McKinnell L.A., 2000, On the predictability of foF2 using neural networks,Radio Science, 1, pp. 225–234.
 
Skone, S.,1998, Wide area ionosphere grid modelling in the auroral region,PhD Thesis, University of Calgary, Department of Geomatics Engineering, Calgary, Alberta, Canada.
Tulunay E., Senalp E.T., Radicella S.M. and Tulanay Y., 2006, Forecasting total electron content maps by neural network technique, Radio Science, 41,doi:10.1029/2005RS003285.
Tulunay, E., Senalp, E.T., Cander, L.R., Tulunay, Y.K., Bilge, A.H., Mizrahi, E., Kouris, S.S., Jakowski, N.,2004, Development of algorithms and software for forecasting nowcasting and variability of TEC. Ann. Geophys. 47 (2/3), 1201–1214.
Watthanasangmechai, K., Supnithi, P., Lerkvaranyu, S., Tsugawa, T., Nagatsuma, T., Maruyama,T., 2012,TEC prediction with neural network for equatorial latitude station in Thailand, Proceeding of the Earth Planets Space, 64, 473–483.
 
 Xenos, T.D., Kouris, S.S., Casimiro, A.,2003, Time­ dependent prediction degradation assessment of neural networks-based TEC forecasting models, Nonlinear Proc. Geophys. 10, 585–587.
Yilmaz,A., Akdogan, K. E., Gurun, M.,2009, Regional TEC mapping using neural networks, Radio Science, 44, RS3007, doi.org/10.1029/2008RS004049.
Meggs, R.,2005, Mapping of ionospheric total electron content using global navigation satellite systems, Ph.D. thesis, University of Bath, Bath, UK.
Misra, P. and Enge, P.,2006, Global Positioning System; Signals, Mea-surements and Performance, Ganga-Jumuna Press, P.O. Box 692 Lincoln, Massachusetts 01773.
Opperman, B. D. L., Cilliers, P. J., McKinnell, L.-A., and Hag-gard, R.,2007,Development of a Regional GPS-based Ionospheric TEC model for South Africa, Advances in Space Research, 39(5),808–815.
Habarulema, J. B., McKinnell, L.-A., and Opperman, B. D. L.,2010,TEC measurements and modelling over Southern Africa during magnetic storms; a comparative analysis, Journal of Atmospheric and Solar Terrestrial Physics, 72(5-6), 509–520.
Perrone L., and Franceschi G. D.,1998, Solar, ionospheric and geomagnetic indices,ANNALI DI GEOFISICA, vol. 41, N. 5-6.
 
Mayaud, P.N., 1980,Derivation, Meaning, and Use of Geomagnetic Indices,Geophysical Monograph 22. Washington, DC: American Geophysical Union.
 
McPherron, R.L.,1995, Standard indices of geomagnetic activity. In Kivelson, M.G., and Russell, C.T. (eds.), Introduction to Space Physics. Cambridge, UK: Cambridge University Press, pp. 451–458.
 
Collins, J. P.,1999, An overview of GPS inter-frequency carrier phase combinations, Geodesy and
Geomatics Engineering Report, University of New Brunswick, pp. 1-15.
 
Schaer, S., Markus, R., Gerhard, B. & Timon, A.S.,1996, Daily Global Ionosphere Maps based on GPS Carrier Phase Data Routinely produced by the CODE Analysis Center, Proceeding of the IGS Analysis CenterWorkshop,Silver Spring, Maryland, pp. 181-192, USA.