Evaluation of Global Ionospheric Maps for Single Frequency PPP in Egypt Amr El-Demiry, Mahmoud Abd Rabbou, and Adel El-Shazly Department of Civil Engineering, Cairo University Presented by; Prof. Adel El-Shazly
Outlines • Introduction • Problem statement • Research objectives • Mathematical models • Results and analysis • Conclusion • Recommendations
Introduction • The ionosphere is part of the Earth’s atmosphere, extending from about 50 km up to about 1,000 km or more above the Earth depending on the sun activities. • The ionosphere layer is a shell of electrons and electrically atoms. • The number electrons affects the GPS signal path through the layer, is affected by the sun activities . • The effect of the ionosphere is proportional to the total number of the electrons along the path of the signals called the total electron content TEC and the signal frequecy. • For GPS signals which have high frequencies, it causes a delay of GPS signal - code modulation and carrier phases. • It causes a delay in code pseudorange measurement and advances the phase measurement by the same amount . http://www.wirelessdictionary.com/Wireless-Dictionary-Ionospheric-Delay- Definition.html
Introduction •
Introduction • The IGS global ionospheric maps (GIMs) are typically used to correct for the ionospheric error for single frequency PPP due to its superior accuracy compared with the empirical methods. • IGS is The main organization responsible for the production of precise satellite ephemerides, clock corrections and ionospheric maps,------ • The products generated from the IGS are based on the combined effort from all, or most of the IGS centers. • Each of the centers uses data collected from all IGS stations around the world. Presently IGS manages a network of 384 stations • For ionospheric maps, four IGS Ionosphere Associate Analysis Centres (IAACs) are contributing their ionospheric products to the IGS. • Include the CODE, ESOC, JPL, and Technical University of Catalonia (UPC). • The four centers produce 2-dimensional ionosphere TEC maps that refer to a 450km shell height. Acquired from IGS website
Problem statement • The accuracy of these maps is affected by the IGS stations distribution and network resolution. • Up to date, unfortunately, there are no IGS stations in Egypt which in role affected the IGS product accuracy for PPP accuracy.
Research Objectives • This research aims to asses the accuracy of the final and rapid GIM ionospheric models for single frequency PPP in Egypt, To do so; • Investigate the traditional accuracy of GIM products for single frequency PPP under well distributed stations network ( North America) to be considered as a reference for these products accuracy in Egypt. • Use GPS dual frequency linear combination mainly the geometry- free to roughly investigate the behaviour of the ionospheric delay effect in Egypt. • Investigate the positioning accuracy and performance of the GIM for SF PPP in Egypt using number of data sets.
Mathematical Model • P = ρ (t,t-τ)+c[dt (t)-dt (t-τ)]+T+I +c[d (t)- d (t- τ)] +m s s 1 +e 1 r 1 r p1 p1 s s 1 = ρ(t,t-τ)+c[dt (t)-dt (t-τ)]+ T - I +c[ (t) - (t- τ)] + m +e 1 r 1 r 1 1 s + [ N + (t )- (t ) ] 1 1 r 1 0 1 0
Mathematical Model •
Mathematical Model • To detect the real behaviour of ionosphere effect leveling geometry- free technique is used. 2 2 f f DCB is the differential code bias 2 1 P P P I DCB e 12 1 2 1 noise 2 f and DPB is the differential phase 2 2 2 bias f f 2 1 I DPB ( N ) 12 1 2 1 12 noise 2 f 2 To apply those equations the frequency depended bias and errors • such as DPB and the ambiguity parameters should be removed from the Geometry free observations To do so, leveling Phase GF observations to the level of the • ionosphere is required to insure that the geometry free contains only the total ionosphere. Code-GF can be used in leveling but in this case the DCB will be added • which worse the accuracy. In this research, GIM ionosphere products are used instead code-GF to investigate the difference between the GIM and real ionosphere variation trend.
Results and Analysis SF PPP Positioning accuracy using final and rapid GIM for North America network. • Data were collected from five North American stations representing different latitudes (low, medium and high). • The data were collected at three different seasons (January, June and October), each for three days of the years: 2008 and 2012to reflect the seasonal variations of the ionospheric delay. • 24 hour data sets with 30 seconds observation interval for the five stations were downloaded from IGS website. • The tropospheric ZPD was using Hopfield model with default atmospheric parameters, and the Neil mapping function is used. • The IGS Final satellite orbit and clock corrections were used.
Results and Analysis • North America stations network. Latitude Longitude Station name (degree) (degree) MDO1 30.68 -104.014 QUIN 39.97 -120.944 PRDS 50.87 -114.293 SASK 52.19625 -106.3983 CRO1 17.75 -64.5843
Results and Analysis • Positioning accuracy for station CRO1 at January 2012 (as an example).
Results and Analysis • Summary of positioning accuracy for the five stations for January 2008 and 2012 (as an example) Jan-12 GIM final rapid station name DLAT(m) DLON(m) DHGT(m) DLAT(m) DLON(m) DHGT(m) cro1 -0.65 0.91 -2.31 0.43 -1.20 1.83 mdo1 0.96 -0.53 0.19 -0.32 0.98 0.76 prds 0.78 -0.64 -0.31 -1.62 0.69 -0.14 quin 0.74 -0.74 1.75 -1.41 0.98 -5.06 sask -0.52 -0.23 -0.42 -0.52 1.26 1.15 Jan-08 GIM final rapid station name DLAT(m) DLON(m) DHGT(m) DLAT(m) DLON(m) DHGT(m) cro1 -0.28 0.07 -0.24 -0.28 0.05 -0.23 mdo1 -0.14 0.49 -0.48 -0.11 0.52 -0.50 prds 0.09 -0.01 -0.70 0.09 -0.09 -0.75 quin 0.24 0.12 -0.84 0.33 0.13 -0.88 sask -0.86 0.52 -1.47 -0.88 0.39 -1.66
Results and Analysis • Estimated ionospheric delay using leveling geometry-free linear combination compared with GIM delay products Comparable vertical ionospheric delay is extracted from GIM for Station CRO1.
Results and Analysis • Estimated ionospheric delay using leveling geometry-free linear combination compared with GIM delay products in Egypt. GPS data at Alexandria GPS data at Helwan In contrast, the GIM ionospheric delay products trend variation is far from the leveling geometry-free ionospheric delay estimated trend.
Results and Analysis • Investigate the positioning accuracy of single frequency PPP in Egypt using final and rapid GIM model. • Three sets of GPS static data were collected in different geographic regions in Egypt mainly Alexandria, Helwan and Aswan. • The data is processed considering IGS precise orbital and clock products. • The tropospheric ZPD was using the empirical Hopfield model with default atmospheric parameters, and the Neil mapping function is used. •
Results and Analysis • Positioning accuracy for Helwan station at July 2014 • It can be seen that no convergence is detected compared with dual frequency results due to the effect of the ionosphere.
Results and Analysis • Positioning accuracy for Alexandria station at April 2009 • Although, the ionospheric effect is eliminated in dual- frequency model, the convergence time extended to 50 minutes due to the clock and orbital products accuracy in weak stations resolution.
Results and Analysis • Positioning accuracy for Aswan station at July 2014 • The single frequency PPP ended up accuracy with 1.5 meter for both latitude and altitude after 4 hours of continues data using final GIM model.
Conclusions • In this research, the performance and accuracy of final and rapid GIM ionospheric model is assessed for single frequency PPP in Egypt. • Well distributed station network is used to represent the traditional accuracy of the GIM ionospheric products as a reference for the accuracy of these products in Egypt. • Leveling geometry-free linear combination (L1-L2) is used to investigate the accuracy of the GIM final ionospheric map products in Egypt in comparison with North America station case. • The results show that the final GIM TEC products failed to detect the actual trend and variation of the ionosphere affected the GPS data in Egypt. As seen from the results, the GIM delay products are affected from the low resolution of IGS stations in Egypt.
Conclusions • In general, both the final and rapid GIM ionospheric model achieve decimeter- level to meter-level accuracy after two hours of continues GPS data processing in well distributed station network such as North America IGS station network. However, the final GIM model present better convergence time than the rapid. • In Egypt, both the final and rapid GIM ionospheric model achieve meter-level accuracy after four hours of continues GPS data processing with no convergence in the positioning solution using three sets of static data at Helwan, Alexandria and Aswan. • The dual frequency ionosphere-free solution present encourage positioning solution compared with the traditional differential positioning solution in Egypt with convergence time extend to 50 minutes.
Recommend
More recommend
