Ionospheric Irregularity Influences on GPS Time Delay

Transcription

1 Ionospheri Irregularity Inluenes on GPS Time Delay Azad A Mansoori 1, Parvaiz A. Khan 1, Shivangi Bhardwaj, Roshni Atulkar and P. K. Purohit * 1 Spae Siene Laboratory, Department o Eletronis, Barkatullah University, Bhopal-4606, India Department o Applied Sienes, National Institute o Tehnial Teahers Training and Researh, Shamla Hills, Bhopal-4600, India ABSTRACT One o most important ionospheri eets on the trans-ionospheri signals is the delay both in range and time. Under this investigation, we have studied the variability o ionospheri range delay in GPS signals. To aomplish this study we have used the GPS measurements at a low latitude station, IISC Bangalore (13.0N, 77.57E) during January 01 to Deember 01. We studied the diurnal, monthly as well as seasonal variability o the range delay. We also seleted ive intense geomagneti storms that ourred during 01 and investigated the variability o delay during these geomagneti storms. From our study we ound the diurnal variability o the range delay is similar to the diurnal pattern observed or TEC. The maximum delay ours during the month o Otober while lowest delay is ound to our in the month o Deember. During summer season the range delay in GPS signals in less while the largest delay ours during the equinox season. The peak delay and enhanement in delay ollows a very good orrelation with Dst index. Key Words: Ionosphere, Range Delay, TEC, Geomagneti Storm 1. INTRODUCTION The GPS signals rom the satellites while propagating through a disturbed ionospheri medium undergo hanges in their harateristis depending on the extent o disturbane. The ionosphere is a dispersive medium, it implies that the ionosphere bends the GPS radio signal rom its optial path and it happens due to hange in its speed while propagating through various layers o ionosphere. A signiiant range error is aused by the hange in the propagation speed. The ionosphere speeds up the propagation o the arrier phase, whereas it slows down the pseudorange ode measurement by an equivalent amount. In other words, the GPS ode inormation is delayed resulting in the pseudorange being measured too long as ompared to the geometri distane o the satellite [Homann et al., 199]. So, the reeiver-satellite distane will be too short i measured by the arrier phase and is too long i measured by ode as ompared to the atual distane. The ionospheri time delay is diretly proportional to the Total Eletron Content (TEC) along the path o propagating signal between the satellite and user (1 meter or 6.15 TEC units on L1 requeny) [Klobuhar et al., 1975]. TEC is highly dependent on many variables suh as loal time, season, geomagneti loation and the level o solar and magneti disturbanes. Strong ionospheri disturbanes have great impat on perormanes o the GPS reeivers. The ionospheri eets on the GPS reeivers have been studied by many researhers [Doherty et al., 1

2 000; Skone, 001; Bhattaharya et al., 009; Jain et al., 010]. The positional auray o the GPS system is limited by the preision in measuring atmospheri time delay. Preise ionospheri time delay estimation is required or ahieving high level o auray in determination o position, navigation and geodesy. It is well established that estimation o preise time delay by means o monitoring the loks on GPS satellites an be limited by the time delay o the earth s ionosphere. At equatorial and low latitudes TEC is highly variable with loal time, season and level o solar and magneti ativity. The dominant variability is diurnal due to the large variation in inident solar radiation, so the time delay is also highly variable at low latitudes. At equatorial regions, the earth s magneti ield is horizontal and there is east-west eletri ield due to the dynami eet produed by the atmospheri motions. During the day the eletri ield is eastward and westward during the night. This phenomenon so auses irregularity in the ionospheri ondition, hene ontribute to the delay mehanism. The range obtained between the satellite and user by integrating the phase and group rerative indies along the path o GPS signal is dierent rom the true range. The dierene between measured range and the true range is known as ionospheri error. This error is negative or the arrier phase pseudoranges and positive or the ode pseudoranges [Komjathy, 1997]. The delay due to the ionosphere results in range errors whih may vary rom ew meters to tens o meters. The ionosphere is a dispersive medium i.e., its rerative index is a untion o the operating requeny [Kaplan, 1996; Mishra and Enge, 006). Thus appropriate methods an be adopted or determining the extent o delay due to ionosphere using ode observations at L1 ( MHz) or at both L1 ( MHz) and L (17.60 MHz) GPS requenies. Typially ionospheri delays on GPS observations an be redued by using the ombination o two broadasting requenies, by using delay model o ionosphere or single requeny users [Kleusberg, 1998]. In reent years various ionospheri delay models were proposed [Klobuhar, 1986; Coster et al., 199]. The eets o ionosphere on GPS perormanes an be onsidered in two aspets: irst during strong ionospheri disturbanes and seond due to the amplitude and phase variations o GPS signals due to the disturbanes, GPS reeiver perormanes are degraded. During these adverse onditions, onventional models an not aurately desribe the ionospheri delay. Thus, or ahieving preise GPS positioning, the ionospheri eets must be eliminated so that the more preise position ould be measured. Hene ionospheri threat models are required to evaluate impat o disturbanes on positioning auray, whih is an important ator or system integrity [Luo et al., 004].. DATA SETS AND METHODOLOGY To aomplish this study we have made use o three types o data sets; GPS data, Dst index and IMF-Bz. A omplete network o GPS reeivers has been setup worldwide sine last ouple o deades, and the observations are arried out regularly. The data obtained in this way is reely available to users. This servie ommonly known as International GPS Servie (IGS) provides the data o hundreds o stations rom all parts o the world. GPS navigation and observation data downloaded rom the IGS stations is in ompressed RINEX ormat. The time samplings o these data are 30 seonds. The TEC along the path rom satellite to reeiver, (STEC), at the two GPS requenies, L1 = 1 = GHz and L = = 1.76 GHz, an be alulated [Klobuhar, 1996]. The GPS TEC data used in this study were obtained rom the IGS or the IGS station IISC Bangalore (13.0N, 77.57E). However, or the present analysis, the data obtained using ode measurement is only used rom January to Deember 01 or all the days. From the proessed data, elevation angle and TEC are used to estimate the time delay values at elevation ut o ESTIMATION OF IONOSPHERIC DELAY

3 The most widely used ionospheri model or estimation o ionospheri delay is the grid based ionospheri model. However we have used another model or estimation o ionosheri delay at user position. This method used GPS pseudo-range measurements at both L1 and L requenies. GPS pseudo-range and arrier phase range measurements are estimated based on assumptions that the signal veloity and wavelength are equal to those values valid or an eletromagneti wave propagating in vauum. However, the ionospheri index o reration has a non-unit value due to the physial properties o the ionosphere, thereore the assumption that the GPS signal travels at the speed o light in vauum and with wavelength equal to the wavelength on vauum is inorret. The group veloity, however, is less than the speed o light, and aused the group delay. The phase and group veloities an be derived as ollows; v v n 40.3N N 1 g g n 40.3N N 1 p p The phase ionospheri range delay, ΔΦ, and the group range delay, ΔP, whih are aused by the phase advane and the group delay, respetively, an thereore be derived by subtrating the assumed veloity,, and the true veloities (v p and v g ) multiplied by the travel time o the signal and an be expressed as ollows: path ( 1) dl Ndl TEC n p path ( 1) dl Ndl TEC P ng path path The magnitude o the range errors is equal or both arrier phase and pseudo range measurements path Ndl but with the opposite sign. The quantity an be evaluated by integrating eletron density along the signal path. This quantity represents Total Eletron Content (TEC). The Total Eletron Content (TEC) is omputed and onverted into ionospheri delay in meters using a onversion ator. Following relation has been used to get the total ionospheri delay (inluding reeiver bias and P1-P bias): TEC 9.483( RL RL 1) TECRC TEC p1 p Where, R L1, is pseudorange at L1 requeny; R L, is pseudorange at L requeny; TEC RC, is reeiver bias error/0.351; and TEC P1-P, is P1-P bias error/0.351, respetively. Thereore, the total ionospheri delay in meters is given as: I TEC Sine the delay due to ionosphere is one o the most important soures o error, in our analysis this delay has been estimated using GPS ode observables and methods using TEC values. Ionospheri orretion terms rom both the methods are applied to the orresponding pseudoranges and user position is estimated. 3. RESULTS AND DISCUSSIONS The ionospheri onditions and so the delay hanges hour to hour, day to day, season to season as well as during disturbed and quiet solar and geomagneti onditions. Thereore, we have studied the variability o ionospheri delay diurnally, monthly as well as seasonally. 3

4 The interplanetary, solar wind and geomagneti onditions during the year 01 are shown in Figure1. The Figure1 shows the variation o Dst index, Kp index, IMF Bz, Solar wind temperature, solar veloity and solar wind density or the year 01. From the Figure we learly notie that there has been a mixed type o ativity during the year 01. There were a number o geomagneti storms some o them intense. Also there was large number o days or whih the geomagneti ativity was quite low. Figure 1: The daily behaviour o various geomagneti and interplanetary indies during the year DIURNAL VARIABILITY The diurnal variation o ionospheri delay or all the days o eah o the twelve months o year 01 is shown in Figure. It an learly be observed rom the igure that the ionospheri delay ollows a diurnal pattern similar to that o TEC. It starts inreasing in the morning o eah day and ahieves peak around 0600 to 100 hrs UT during all months o the year 01. The delay reorded highest peaks during the months o April, September and Otober with peak values o about 14 meters while the shallow peaks were observed during the month o June, July, Deember, January and February with peak values o about 8 meters. The diurnal pattern observed during all the months has same shape with ourrene o diurnal peak around the same times. 4

5 Figure : The diurnal variability o the ionospheri delay during all the months o the year MONTHLY VARIABILITY The month to month variability o the ionospheri range delay or eah month o year 01 is shown in Figure 3. The variability during all the days o eah month is averaged to onstrut the Figure 3. The igure shows that the monthly variation o ionospheri delay is highest during the month o Otober and reahes a value o 6 meters while the least delay is observed during the month o Deember with value o 3.6 meters. The ionospheri delay starts inreasing rom the month o Deember and ahieves peak in the month o Marh ater that it again starts dereasing and reahes minimum in the month June or July. We also notie that the monthly variability ollows semi-annual variability. Figure 3: The diurnal, monthly and seasonal variability o the ionospheri delay during the year 01. 5

6 3.3. SEASONAL VARIABILITY We have also studied the seasonal variability o ionospheri delay at IISC Bangalore. The seasonal variability o ionospheri delay during three dierent seasons o the year 01 at IISC Bangalore is shown in Figure 3 (bottom panel). The igure shows that the ionospheri delay is maximum during the equinox season with peak value o 5.5 meters while the minimum delay is observed during the summer season with peak value 4. meters GEOMAGNETIC VARIATION We then took the peak values o Dst o all the ive storms and the peak enhanement in ionospheri delay during eah storm to aess the magnitude o orrelation between storm intensity index and the ionospheri delay. The satter plot o peak Dst with enhanement in ionospheri delay during all the ive storm events is shown in Figure 4. From the igure we notie the satter between enhanements produed in the ionospheri delay during eah storm with the orresponding intensity o that storm is not muh large. We alulated the orrelation oeiients o peak ionospheri delay and the enhanement in ionospheri delay with the storm intensity index Dst. We ound that both peak values and enhanement in ionospheri delay exhibit a moderate orrelation with the storm intensity index, Dst with orrelation oeiients 0.60 and 0.65 respetively. Figure 4: Correlation o storm intensity index, Dst with peak ionospheri delay and enhanement in ionospheri delay. 4. CONCLUSIONS The ionospheri delay ollows a typial diurnal pattern ahieving a normal diurnal peak around 06:00 to 1:00 UT during all the months o the year. The maximum ionospheri delay is observed during the month o Otober while the minimum delay is observed during the month o Deember. The maximum delay is observed during equinox season while the minimum delay is observed during the summer season. 6

7 The ionospheri delay is strongly aeted during the disturbed geomagneti onditions. During all the seleted ive geomagneti storm events we ound a positive enhanement in ionospheri delay. A good orrelation exists between the peak values o Dst and ionospheri delay as well as between the peak values o Dst and enhanement in ionospheri delay. ACKNOWLEDGEMENTS The author is thankul to UGC, New Delhi or providing inanial assistane through MANIF ellowship programme. The data taken rom SOPAC data server is also highly aknowledged. REFERENCES Bhattaharya, S., Purohit, P. K., & Gwal, A. K. (009), Ionospheri time delay variation in the equatorial anomaly region during low solar ativity using GPS, Ind. J Rad. Sp. Phys., 38, Coster, A. J., Gaposhldn, E. M., & Thornton, L. E. (199), Real-time ionospheri monitoring system using GPS, J. Inst. Navig., (USA), 39(), Doherty, P. H., Delay, S., Valladares, C., & Klobuhar, J. (000), Ionospheri sintillation eets in the equatorial and auroral region, Proeeding o ION-GPS 000, Salt Lake City, Utah, Homann-Wellenho, B., Lihtenegger, H., & Collins, J. (199), Global Positioning System: Theory and Pratie, 4 th edition, Springer-Verlag, Berlin Heidelberg New York. Jain, A., Tiwari, S., Jain, S., & Gwal, A. K. (010), TEC response during severe geomagneti storms near the rest o equatorial ionization anomaly, Ind. J Rad. Sp. Phys., 39, Kaplan, E. D. (1996), Understanding GPS: Priniples and Appliations, Norwood, MA, Arteh House. Kleusberg, A. (1998), Atmospheri model rom GPS, GPS or geodesy, nd Edition, Springer, Wien. Klobuhar, J. A. (1975), Polarization o VHF waves emitted rom geostationary satellites, J. Geophys. Res., 80(31), Klobuhar, J. A. (1986), Design and Charateristis o the GPS ionospheri time delay algorithm or single requeny users, Proeeding o the PLANS-86 onerene, Las Vegas, NY, Klobuhar, J. A. (1996), Ionospheri Eets on GPS, in Global Positioning System: Theory and Appliations, Volume 1, ed. by B. W. Parkinson and J. J. Spilker, Amerian Institute o Aeronautis and Astronautis, Washington, DC. Komjathy, A. (1997), Global Ionospheri Total Eletron Content Mapping Using the Global Positioning System, Ph.D. dissertation, Department o Geodesy and Geomatis Engineering Tehnial Report No. 188, University o New Brunswik, Frederiton, New Brunswik, Canada, 48pp Luo, M., Pullen, S., Ene, A., Qiu, D., Waller, T., & Enge, P. (004), Ionosphere threat to LAAS: Updated model, user impat and mitigations, ION-GNSS 004, Long Beah, CA, Mishra, P., & Enge, P. (006), Global Positioning System: Signal, Measurements and Perormane, nd edition, Ganga Jamuna Press, Massahusetts. Skone, S. H. (001), The impat o magneti storm on GPS reeiver perormane. J. Geodyn, 75(6),