Home

Halo_GPS software user manual

Contents

1. ref_i_start_epoch_num sat_id 1 ref_i_start_epochnum sat_id k ref_i_end_epochnum sat_id k ref_i_common_sat_num epoch 1 ref_i_common_satnum epoch 1 ref_i_change_sat epoch 1 ref_i_lost_sat_num epoch 1 ref_i_lost_satnum epoch i ref_i_add_sat_num epoch 1 ref_i_add_satnum epoch i 0 no change 1 lost satellite 2 add satellite 3 lost satellite and add satellite This subroutine searches the common satellites between reference station and kinematic station And all information of satellite s changing is saved det_cycle_slip f90 Input parameters Reference station ID Kinematic station ID Output parameters The number of cycle slips of every satellite The epoch number of cycle slips of every satellite The number of outliers of every satellite 11 Scientific Technical Report STR 10 11 10 2312 GFZ b103 10119 ref_i_slip_num sat_id k ref_i_slipnum sat_id k kk ref_i_outlier_num sat_id k Deutsches GeoForschungsZentrum GFZ The epoch number of outliers of every satellite ref_i_outliernum sat_id k kk Introduction The subroutine is used to detect the cycle slip and outlier of every satellite based on the median method And it removes those observations if the observation time is less than the minimum observation time At the same time the information of satellite rising and downing is updated Additionally the reference satellite is chosen automatically
2. 7 Initialization of matrix and all unknown parameters 8 Parameters estimation using sequential adjustment 9 Robust estimation 10 Output the results and summary 11 End of Program Read User Command File Read Ref Station Obs Files Read Navigation Files Read Kin Station Obs Files T 7 I I y Obtain Ref Station Obs Infor Obtain Kin Station Obs Infor Obtain Other Obs Infor l I T y Data Fusion y Data Pre processing y Form Double Difference Obs y Initial All Unknown Parameters and Matrix y Adjustment Filter y Analysis of Residuals y Put Out Results Fig 1 The data processing flowchart of HALO_GPS Read Other Files 13 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 3 Control File The essential work to run the program for GPS data processing is to write an input parameter file defined in a flexible form Definitions of the input parameters and control file format as well as a standard control file template are introduced in following subsections 3 1 Definitions of Input Parameters 1 Address of precise ephemeris character len 57 address_sp3_file 2 Address of broadcast ephemeris character len 57 address_brdc_file 3 Address of observation file at the 1st reference station ch
3. HELMHOLTZ ZENTRUM POTSDAM Helmholtz Zentrum DEUTSCHES POTSDAM GEOFORSCHUNGSZENTRUM Qianxin Wang Tianhe Xu Guochang Xu HALO_GPS Software User Manual High Altitude and LOng Range Airborne GPS Positioning Software Version of 2010 Scientific Technical Report STR10 11 A HELMHOLTZ www gfz potsdam de GEMEINSCHAFT Impressum HELMHOLTZ ZENTRUM POTSDAM DEUTSCHES GEOFORSCHUNGSZENTRUM Telegrafenberg D 14473 Potsdam Gedruckt in Potsdam November 2010 ISSN 1610 0956 Die vorliegende Arbeit in der Schriftenreihe Scientific Technical Report STR des GFZ ist in elektronischer Form erh ltlich unter www gfz potsdam de Neuestes Neue Publikationen des GFZ Qianxin Wang Tianhe Xu Guochang Xu HALO_GPS Software User Manual High Altitude and LOng Range Airborne GPS Positioning Software Version of 2010 Scientific Technical Report STR10 11 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 HALO_GPS High Altitude and LOng Range Airborne GPS Positioning Software Software User Manual Version of 2010 Qianxin Wang Tianhe Xu Guochang Xu GFZ German Research Centre for Geosciences Department 1 Geodesy and Remote Sensing Telegrafenberg 14473 Potsdam Germany June 2010 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 HALO GPS Software User Manual Contents De INtrOGUCE ON ia o a A AA A ARA 3
4. TT OTULS
5. 7 Then we use HALO_GPS software to process these data during the antenna moving Fig 8 shows the positioning results of HALO_GPS The calculation results agree well with the true increments of antenna 19cm Fig 7 The vertical motion experiment 30 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 0 30 60 90 120 150 180 210 Epoch 30 seconds Fig 8 The positioning results on the height component Another is horizontal motion test Similarly the IGS reference station of GFZ is taken as the fixed station But the antenna of kinematic station is placed on a rule The initial location of antenna on the rule was 103 cm After half an hour the antenna was moved to 43 cm on the rule see Fig 9 Fig 10 shows the positioning results of HALO_GPS Fig 9 The horizontal motion experiment 0 8 The Distance gt a 0 4 Distance meters gt iv 0 40 80 120 160 200 Epoch 30 seconds Fig 10 The positioning results of motion distance on the horizontal component Fig 10 shows the calculation results agree well with the true distance of antenna moving 31 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 7 3 Sea Buoy Experiment This experiment was performed by Hong Kong Polytechnic University on December 8th 2004 at Repulse Bay Hong Kong Island Two Leica dual frequency GPS receivers were used one is set on sh
6. Rio de Janeiro Brazil 1997 Wackernagel H Multivariate geostatistics an introduction with applications Springer Verlag 291 292 1998 Wang C H Liou Y A Yeh T K Impact of surface meteorological measurements on GPS height determination Geophysical Research Letter 35 L23809 2008 Wang Qianxin Xu Guochang S Petrovic U Schaefer U Meyer Xu Tianhe A regional tropospheric model for 38 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 airborne GPS applications Advances Space Research 2010 under review Wang Qianxin Xu Guochang Chen Zhengyan Interpolation method of tropospheric delay of high altitude rover based on regional GPS network Geomatics and Information Science of Wuhan University Vol 35 11 2010 Wang Qianxin Xu Tianhe Xu Guochang Application of robust estimation to GPS airborne kinematic relative positioning Geomatics and Information Science of Wuhan University 2010 under review Wang Qianxin Xu Tianhe Xu Guochang GPS kinematic positioning for long range flight based on single baseline model Acta Geodaetica et Cartographica Sinica 2010 submitted Wang Qianxin Xu Guochang Real time GPS Satellite Clock Error Prediction Based On No stationary Time Series Model Europe Geosciences Union EGU Vol 11 EGU2009 5993 1 19th 24th April Vienna Austria Wang Qianxin Li Li Gong Youxing Study of GPS satellite clock s behaviors and prediction Science of
7. Note The reference satellite is always placed on the Ist order in HALO_GPS form_sd_dd 90 Input parameters Reference station ID Kinematic station ID Output parameters The single differenced LC observations ref_1_vobs8_sd epoch sat_id The double differenced LC observations ref_1_vobs8_dd epoch sat_id The initial double differenced ambiguity of LC observations ref_i_dd_amb_LC sat_id k Introduction The subroutine is used to compute the single and double differenced LC observations Additionally the initial double differenced ambiguity of L1 L2 and LC observation is fast obtained by a new method which is introduced in Chapter 5 a_x_1 90 Input parameters Reference station ID Output parameters A_row A_rank the row and column of matrix_A X_row X_rank the row and column of matrix_X L_row L_rank the row and column of matrix_L P_row P_rank the row and column of matrix_P ref_i_amb_num the number of ambiguity paramters ref_i_xyz_num the number of position paramters logical_amb j sat_id k the Kth ambiguity of ith satellite at jth epoch Introduction This subroutine computes the row and column number of some important matrix which will be used in the adjustment sequential_adjustment f90 Input parameters Reference station ID the structure variable IGS Output parameters The final coordinate of kinematic station kin_i_X epoch 1 kin_i_Y epoch 1 kin_i_Z epoch 1
8. are connected with the main function main f90 via the input and output parameters Therefore it is very important for using this software to known the definitions of input parameters output parameters and the function of each subroutine 2 1 Main Function The main function is the most important part in the most of software which is used to organize the rest of the subroutines The main function of HALO_GPS includes 15 important steps which are outlined below Reading Control File Reading Precise Ephemeris Reading Observation File of Reference Station Reading Observation File of Kinematic Station Scanning and Modifying All Observation Data Single Point Positioning Computing Receiver Clock Errors Searching Common Satellites RAR AS Cycle Slip Detection Choosing Reference Satellite Ra Forming Double Observation Equations N Initializing Matrixes and Ambiguities W Parameters Estimation with Sequential Adjustment H A Scanning the Residuals ha nn Generating Result Files and Summary 2 2 Important Subroutines read_control_file f90 Input parameters The address of control file Example Example control_file model_control_file_1 txt Output parameters A logical variable which represents whether the address of control file is right or not Introduction This subroutine is used to read the control file defined by the user in a flexible form Some important
9. differential GPS positioning Proceedings of ION GPS 2005 179 188 Long Beach California 2005 39 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 11 Appendixes 11 1 Appendixes 1 Definitions of Constants This section describes all of the constants which are used in HALO_GPS software Part 1 introduces definitions of all constants name Part 2 gives the special values of all constants used Part 1 earth_flat earth_rad earth_to_moon g earth GM_moon GM_sun GM_earth G_univ pi rad_to_deg rad_to_mas sec_per_day Earth s flattening Equatorial radius of the Earth m Mass ratio of earth and moon Gravitational acceleration at the equator m s 2 GM for moon GM for sun GM for Earth Gravitational constant Pi Conversion from radians to degrees Conversion from radians to milliarcseconds Number of seconds in 24 hours vel_light speed of light in m s DJ2000 Julian date of J2000 sec360 number of seconds in 360 degrees solar_to_sidereal Conversion from solar days to sidereal days at J2000 fL1 fL2 GPS frequencies in Hz at L1 and L2 dfsf sfdf Difference of frequency divided by the sum of frequencies used form widelane and narrowlane Icf1 lef2 Multipliers for LC from L1 and L2 frequencies Igfl lgf2 Multipliers for LG from L1 and L2 frequencies pefl pef2 Multipliers for PC
10. distance which is defined by user the new reference station will used to replace the old one At the same time all information of old observation equation including covariance matrix are transferred to the new observation equation based on the equivalent eliminated parameter method The calculation steps of adaptively changing reference station are described below Firstly we suppose that the observation equation before changing reference station can be written as x L A X V P 13 VAN i And the observation equation after changing reference station can be expressed as X L B X V P 14 VAN a 52 where L L are the observations A B are the design matrices X X X X are the position parameters VAN VAN are the double differenced ambiguities between old reference station il and kinematic station i2 and those of new reference station i3 and kinematic station i2 respectively V V P P are the residual vectors and the weight matrices respectively It is to be noted that X X are the same position parameters of kinematic station Then the Eq 13 can be rewritten as L A 4 V P 15 X where X includes X and VAN Normal equation of Eq 15 can be obtained as le EIA 16 M M X U where 25 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 M M _ APA A PA 17 M Mn A PA A PA U _ APPL 18 U A PL T
11. from P1 and P2 frequencies Part 2 parameter earth_flat 0 003352891869D0 parameter earth_rad 6378145 DO parameter earthrot 7 29212E 05 parameter earth_rad 6378137 DO parameter earth_flat 1 d0 298 257222101 parameter earth_to_moon 81 30065918D0 parameter g earth 9 780318458D0 parameter GM_moon 0 49027975D 13 parameter GM_sun 0 132712499D 21 parameter GM_earth 3 986004418d 14 Scientific Technical Report STR 10 11 10 2312 GFZ b103 10119 40 Deutsches GeoForschungsZentrum GFZ parameter G_univ 0 66732D 10 parameter pi 3 1415926535897932D0 parameter sec_per_day 86400 D0 parameter sec360 1296000 d0 parameter vel_light 299792458 D0 parameter DJ2000 2451545 d0 parameter solar_to_sidereal 1 002737909d0 parameter fL1 154 10 23d6 parameter fL2 120 10 23d6 parameter wavel vel_light fL1 parameter wave2 vel_light fL2 parameter wave_LC vel_light fL1 parameter wave_WL vel_light fL1 fL2 parameter rad_to deg 180 d0 pi parameter rad_to_mas 648000 d3 pi parameter dfsf fL1 fL2 fL1 fL2 parameter sfdf L1 fL2 L1 fL2 parameter lcfl 1 d0 1 d0 fL2 fL1 2 parameter lcf2 fL2 fL1 1 d0 fL2 fL1 2 parameter Igfl fL2 fL1 parameter lgf2 1 d0 parameter pefl fL1 2 fL1 2 fL2 2 parameter pcf2 fL2 2 fL1 2 fL2 2 11 2 Append
12. gravimetry campaign And the internal tests and external comparisons are also made Fig 13 shows the plane trajectory of airplane in the NorthGrace2007 campaign This campaign includes a total of 25 flights The positioning results of two flights on June 10th 2007 and June 14th 2007 are shown here Fig 14 is the comparisons between HALO_GPS and GAMIT on the height component dd Fig 13 The plane trajectory of 25 flights in NorthGrace2007 campaign 07 June 10th 2007 HALO GAMIT June 14th 2007 HALO_GAMIT S 2 S E FERN on f I 5 5 S E 1 1 dH meters dH meters tx a 1 ES ato 5 ES l Mean 0 152 Mean 0 156 Std 0 018 1 Std 0 015 0 5 7 1 r 7 r T 0 3600 7200 10800 0 3600 7200 10800 Epoch seconds Epoch seconds Fig 14 The comparison of the positioning results between HALO_GPS and GAMIT Fig 14 shows there is a 15 cm bias between the positioning results of HALO_GPS and that of GAMIT The reason may be that the different error correction models are used in the different software However the standard deviations are better than 2 cm The results are satisfied 7 5 AlpinAero2008 Campaign The AlpinAero2008 was an airborne survey in the Alps and their German forelands carried out by 33 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 BKG in cooperation with the GFZ Potsdam and BGR Hanover in September October 2008 u
13. is treated as a kinematic station to estimate its coordinates at every epoch The tropospheric delays are obtained from remove restore method RRM and PEM respectively The observation time is from 00 00 00 to 06 59 59 on 17th Dec 2008 GAMIT daily solution is regarded as reference values Fig 2 and Fig 3 show the residuals of tropospheric delays and the residuals of estimated elevations using RRM and PEM respectively Table 1 presents statistical results of kinematic positioning precision 21 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 157 ta RRM RMS 2 3 cm PEM RMS 1 5 cm RRM Residuals mm Residuals meters 9 T T T T T T 1 3 Time hours Time hours Fig 2 The residuals of tropospheric delays Fig 3 The residuals in height component using RRM and PEM using RRM and PEM Table 1 Comparison of the precision of GPS kinematic positioning using RRM and PEM unit cm Method L B H RRM rms 0 83 1 37 2 32 PEM rms 0 85 1 34 1 51 Fig 2 and Fig 3 show a negative correlation between ZTDs and the elevation When the estimated ZTDs are larger than the true ZTDs the estimated elevation will be smaller than the true elevation when the estimated ZTDs are smaller than the true ZTDs the estimated elevation will be larger than the true elevation From Table 1 the positioning precisions on two horizontal components are similar using RRM and PEM However the
14. kin_i_B epoch 1 kin_i_L epoch 1 kin_i_H epoch 1 The precision information at every epoch kin_i_sigmaX epoch 1 kin_i_sigmaY epoch 1 kin_i_sigmaZ epoch 1 kin_i_sigmaB epoch 1 kin_i_sigmaL epoch 1 kin_i_sigmaH epoch 1 12 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 RMS at every epoch kin_i_RMS epoch 1 The final double differenced ambiguity of LC observation ref_i_dd_amb_LC sat_id k Introduction The final coordinate and ambiguities of kinematic station are calculated based on the robust sequential adjustment 2 3 Diagram of Software The data processing steps of this software are described in Fig 1 1 Program start 2 Read input parameter file for controlling the run of the software an example of the definition of the input parameter file is presented in Chapter 3 3 Read all possible data files for the run of the software e g satellite information file station information file tropospheric delay correction file receiver antenna phase centre correction file etc 4 Compute all possible corrections e g antenna phase centre correction earth tide correction tropospheric delay correction clock error offsets etc 5 Data preprocessing e g searching the common satellite detecting cycle slip and outlier removing bad observation and receiver clock jump etc 6 Construction of single and double differenced observations
15. matrix transformations we get the general form of factor matrix C C BA AA 12 Finally the new double differenced ambiguity VAN is easy to be obtained by the old double differenced ambiguity VAN multiplied by the matrix C 5 8 Automatic Choosing and Changing Reference Station This function is specially developed for HALO project Although the principle and method are 24 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 introduced systematically in our publications the theories will be described briefly here The single baseline model is the simplest and commonly used model in kinematic relative positioning However such a model is generally not suitable for the long range flight positioning Due to the long distance between reference station and kinematic station many kinds of common errors can not be cancelled out by the difference method And the number of common satellites will be decreased with the increase of baseline length If a closer reference station can be used in place of the original reference station these problems will be solved well Therefore a method of adaptively changing reference station for long distance airborne GPS applications is developed in HALO_GPS software The basic idea is that the positioning model always keeps the single baseline model during the whole solution When the distance between kinematic station and reference station is longer than the maximum
16. of read epoch L1 L2 C1 P1 P2 C2 D1 D2 S1 S2 LI MW LC Additional function Remove the abnormal observations ref_ 1 Hant ref 1 Eant ref_i_Nant tef_i_obstype_num ref_i_obstype 10 ref_i_interval ref_i_G_time epoch 1 tef_i_G_time_MJD epoch 1 ref_i_G_sat_num epoch 1 ref_i_satnum epoch 32 ref_i_read_epoch tef_i_vobs1 epoch sat_id tef_i_vobs2 epoch sat_id tef_i_vobs3 epoch sat_id tef_i_vobs4 epoch sat_id tef_i_vobs5 epoch sat_id tef_i_vobs21 epoch sat_id tef_i_vobs22 epoch sat_id ref_i_vobs23 epoch sat_id tef_i_vobs24 epoch sat_id ref_i_vobs25 epoch sat_id tef_i_vobs6 epoch sat_id tef_i_vobs7 epoch sat_id tef_i_vobs8 epoch sat_id Remove the observations of those satellites which belong to question satellites Remove the observations of GLONASS satellites Scientific Technical Report STR 10 11 10 2312 GFZ b103 10119 Deutsches GeoForschungsZentrum GFZ read_kin_obsfile f90 Input parameters Address of observation file kinematic station ID start time the number of epochs Output parameters It is similar to read the observation file of reference station earth_tide f90 Input parameters Julian Day XYZ coordinates of station Output parameters Tidal correction to site position m Introduction The subroutine is used to compute the solid earth tide based on the formulations in DSR thesis with extensio
17. variables are assigned initial values based on the definitions in the control file The definitions of control file will be described in Chapter 3 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 read_sp3 f90 Input parameters The address of precise ephemeris Output parameters The number of healthy satellites and their ID The number of lost satellites and their ID The number of bad satellites and their ID The bad satellite means the satellite clock error valued is 999999 999999 All information is saved in the structure variable named as igs igs epoch Rtime wenn seconds at this epoch igs epoch RMJD wenn Modified Julian Day at this epoch igs epoch Rx satellite_id 4 4 4 X coordinate of this satellite at this epoch igs epoch Ry satellite_id 4 4 4 4 4 4 2 m Y coordinate of this satellite at this epoch igs epoch Rz satellite_id Z coordinate of this satellite at this epoch igs epoch Rclock satellite_id clock errors of this satellite at this epoch igs epoch Rlogical satellite_id health of this satellite ymdhms_to_MJD f90 Input parameters Year Month Day Hour Minute Second Output parameters Modified Julian Date MJD Introduction The routine converts a calendar date with hour minute and second to a Modified Julian date The calendar date is ordered as year month day hour minute and second These values are stored in a s
18. 0 Input parameters Reference station ID the structure variable IGS Output parameters 10 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 The tropospheric delay of reference station at every epoch ref_clock_3 f90 Input parameters Reference station ID the structure variable IGS Output parameters The clock errors of reference station Satellite elevation angle at each epoch The weight of observations Introduction ref_i_trop epoch sat_id ref_i_C epoch 1 tef_i_elev epoch sat_id ref_i_sat_p epoch sat_id This subroutine computes the clock error of reference station using Pseudorange C1 known site coordinate and the weight matrix of ref_clock_1 solution again In the same time it deletes those satellites with the low elevation angle search_common_sat f90 Input parameters Reference station ID Kinematic station ID Output parameters The total number of common satellites All common satellite IDs The times of each satellite in view The start epoch of the Kth time in view The end epoch of the Kth time in view The number of common satellites at each epoch Common satellite IDs at each epoch Logical variable of the satellite changing The number of lost satellites Lost satellite ID The number of added satellites Add satellite ID The type of satellite changing Introduction ref_i_tot_common_sat_num 1 1 ref_i_tot_common_satnum 1 1
19. 0 0 023 2417327 075 0 014 0 029 7 3 2010 1 2 0 0 10 00 2399062 733 0 011 5389237 874 0 029 2417327 087 0 018 0036 7 4 2010 1 2 0 0 15 00 2399062 735 0 009 5389237 876 0 023 2417327 082 0014 0 028 7 5 2010 1 2 0 0 20 00 2399062 731 0 010 5389237875 0 025 2417327 080 0015 0031 7 19 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 5 Strategies and Principles The HALO_GPS is developed to fulfill the needs of German HALO High Altitude and LOng Range Research Aircraft project The main strengths of the proposed HALO aircraft are its long range about 10000 km and endurance more than 10 flight hours high ceiling altitude more than 15 km and large instrument load capacities which are not available in such combination on any other research aircraft in Europe Therefore it brings many new challenges for airborne GPS kinematic positioning In order to obtain a precise aircraft trajectory some new strategies and techniques are developed in HALO_GPS Although these methods are introduced in the references the theories will be described briefly 5 1 Outlier and Cycle Slip Detection An ambiguity break a so called cycle slip occurs when the phase observation jumps by a few or more cycles in L1 frequency L2 frequency or in both A single outlier is defined as an isolated shift of observation which lasts only one or a few epochs whereas a cycle slip causes a systematic constant shift in th
20. 2 Structure OL software a RR RAN DER Ea discado ceca 4 2 11Main Function ee A A AAA id 4 2 2 Important SUDLOUI iii ida seat boabsleducessesesnsbedddueceselevaseeldducenes 4 2 3 Diagrami OL Software nnd ana rd deb peri dei ctrl dies 13 3 Control A A dabdedslebecuoedeseavodagbacueenssdubbads i a ian 14 3 1 Definitions of Input Parameters jco niia i a EE A E A a 14 3 2 Control Pile Format iia Ad 15 3 3 An Exampleof Control Fil uo dp ee 15 4 File Format han are ar e e aed a 19 4 1 Input Fil Format au ea paa sed RE AE AE E E AE A NAR 19 4 2 Output File Formatare seinna anr a aa 19 3 Strategies and Principles une rin Hr 20 5 1 Outlier and Cycle Slip Detection uussessesnesnessersnesnennennonsnnnnennonnonnnesnennonnannnenennennan 20 3 2 Clock Error Estimation sp 2 2 2822 2232 alain aos 21 5 3 Tropospherie Delay Corrections anne da ta 21 5 4 Ambiguity Resolution 2000sesseesersnesnennnennrsoesnensnesnrsnesnennnsassnennnnsnsssennensnrsnsssssnnnnen 22 DD Rob st ESO voii ii a RER id li a eaan a as 23 A AA O AAE TAEA a SENT 24 5 7 Automatic Choosing and Changing Reference Satellite neeeesesnesnensensnennennenennn 24 5 8 Automatic Choosing and Changing Reference Station ooooncnncnocnncnnoncnncononnnonacanononincnncnnos 24 6 Run of HALOGGPS 222 en en ee gets npaehh era 28 T Numerical Fxamplesi 2 2 3228 18a il RR 29 7 1 Static Data Kinematic Processing ooccoccoccnononnconcnnnonononon
21. 4320 5760 7200 8640 10080 11520 12960 14400 15840 unit meters 241732730 4 2417327 20 4 a ne Positioning Precision 2417327 10 4 unit meters J X Y Z unao TRUE 2399062 746 5389237 882 2417327 073 y Mean 2417327 074 Solu 2399062 753 5389237 891 2417327 074 173 0 A eo Difference 0 007 0 009 0 001 0 1440 2880 4320 e pa pa ii ma mm i mm Epoch 5 seconds 29 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 Fig 5 The results of static station kinematic processing Fig 5 shows the differences are the mm level between the means of HALO_GPS and the true values of IGS daily solution in X Y Z three components And the standard deviations are 1 3 cm 7 2 Antenna Movement Experiment For testing the capability of processing kinematic data the antenna movement experiment is carried out on the roof of building A17 located at GeoForschungsZentrum on February 19th 2010 One is vertical motion test In this test the IGS reference station of GFZ is taken as the fixed station And another GPS station is treated as the kinematic station nearby this IGS reference station which is set up by the Department 1 1 of GEZ see Fig 6 F y Reference station Kinematic station Fig 6 Used IGS reference station and kinematic station at GFZ The initial antenna height of kinematic station was 22 5 cm After half an hour the antenna height was increased to 41 5 cm see Fig
22. Surveying and Mapping No 2 2010 Xu Guochang P Schwintzer and Ch Reigher KSGSoft Kinematic Static GPS Software Software User Manual Scientific Technical Report 19 1998 GeoForschungsZentrum Germany 1998 Xu G A concept of precise kinematic positioning and flight state monitoring from the AGMASCO practice Earth Planet Space 52 831 835 2000 Xu Guochang GPS Data Processing With Equivalent Observation Equations J GPS Solutions 2002 6 28 33 Xu Guochang GPS theory algorithms and application 2 Ed Springer 232 2007 Xu Tianhe Yang Yuanxi The hypothesis testing of scale parameter in coordinate transformation model Geomatics and Information Science of Wuhan University 26 70 74 2001 Yang Yuanxi Robust Estimation for Dependent Observations J Manuscripta Geodaetica 1994 19 1 10 17 Yang Yuanxi Zeng Anmin Fusion Modes of Various Geodetic Observations and Their Analysis J Geomatics and Information Science of Wuhan University 2008 33 8 771 774 Yin H Huang D and Xiong Y Regional tropospheric delay modeling based on GPS reference station network VI Hotine Marussi Symposium on Theoretical and Computational Geodesy 132 185 188 2008 Zhang J Lachapelle G Precise estimation of residual tropospheric delays using a regional GPS network for real time kinematic applications Journal of Geodesy 75 255 266 2001 Zheng Y Feng Y Interpolating residual zenith tropospheric delays for improved regional area
23. The process strategy of ambiguity solution character len 57 amb_pro_method 14 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 54 The tropospheric delay file from user character len 57 address_user_ZTD_file 55 The ephemeris file from user character len 57 address_user_EPH_file 56 Satellite elevation cut off angle real 8 min_elev 57 Minimum observation time integer 4 min_lag 58 Adjustment method character len 21 adjust_method 59 Robust estimation switch integer 4 robust 60 Definitions of parameters in robust estimation real 8 robust_k0 robust_k1 61 The method of weight determination integer 4 elev_p 62 Address of result file 1 Character len 57 address_result_file_xyz 63 Address of result file 2 Character len 57 address_result_file_blh 3 2 Control File Format The proper control file format is necessary to run HALO_GPS software There are a total of 63 command lines in the control file Each line includes 100 characters where the front of 60 characters is the commands and the back 40 characters are the comments The 1st character of each line is a switch where O means close and 1 is open Example Switch commands 60 characters 11 Example data_file AlpinAero2008 igs14984 sp3 3 3 An Example of Control File comments 40 characters ISp3 file 1 The following is an example of a standard c
24. al adjustment algorithm is especially suitable for kinematic case to separate the time dependent unknowns and time independent unknowns so one can solve for position every epoch in one hand and obtain the updated ambiguity information for further use in the other hand The estimated coordinate will be improved as the ambiguity information accumulated The best ambiguity solution will be obtained when the whole observations involved so a repeat computation of the kinematic coordinates by using the best known ambiguity is necessary for the homogeneous coordinate solutions 5 7 Automatic Choosing and Changing Reference Satellite In the long time GPS surveying changing reference satellite is inevitable Therefore we develop an efficient method to deal with this problem It is introduced below The relationship between un differenced ambiguity and double differenced ambiguity before changing reference satellite can be expressed as VAN AN 9 And the relationship between un differenced ambiguity and double differenced ambiguity after changing reference satellite can be expressed as VAN BN 10 where VAN VAN are the old and new double differenced ambiguity respectively A B are the transformed matrix N is un differenced ambiguity The relationship between the new double differenced ambiguity and the old double differenced ambiguity can be assumed as VAN CVAN 11 Therefore it is key problem that how to get the matrix C Through some
25. al ionospheric conditions Thus this combination of ambiguities will be close to constant In our experiment the average variation size of the extra widelane ambiguities is around 0 02 to 0 2 cycles for most of satellites 20 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 For the same reason W M combination can be applied which introduces noise from pseudorange measurements but eliminates ionospheric effects The RMS of widelane ambiguity in Eq 2 is usually 0 2 0 5 cycles The treatment of the cycle slips includes two steps detection and fixing Firstly we examine the ambiguity N6 of widelane L6 and ambiguity N4 of extra widelane L4 The widelane ambiguity N6 is ionosphere free and suffers only from system measurement noise and multipath variations Also the ionospheric effects are largely reduced in L4 Such errors should be smooth over several epochs compared to a cycle break When a discontinuity occurs in the widelane there is a possible cycle break there Then we decide these doubtful cycle slips based on a statistical test model The median method is used to avoid the detection is broken down by some too larger abnormal observations 5 2 Clock Error Estimation The clock error is one of the major errors in GPS surveying A small clock error may cause a very large code and phase error Therefore we have to carefully model the clock error on the satellites and receiver We us
26. aracter len 57 address_ref_1_obsfile 4 _ 11 Address of observation file at the 2nd 9th reference stations 12 Address of observation file at the 1st kinematic station character len 57 address_kin_1_obsfile 13 Name of the Ist reference station character len 57 name_ref_1 14 _ 21 Name of the 2nd 9th reference stations 22 Name of the Ist kinematic station character len 57 name_kin_1 23 XYZ coordinates of the Ist reference station real 8 X_ref_1 Y_ref_1 Z_ref_1 24 _ 31 XYZ coordinates of the 2nd 9th reference stations 32 XYZ coordinates of the Ist kinematic station real 8 X_kin_1 Y_kin_1 Z_kin_1 33 The start time of data processing at the 1st reference station integer 4 ref_1_start_year integer 4 ref_1_start_month integer 4 ref_1_start_day integer 4 ref_1_start_hour ref _1_start_min real 8 ref_1_start_sec real 8 ref_1_start_MJD 34 The specified epoch number at the 1st reference station integer 4 ref_1_num_epoch 35 _ 50 The start time of data processing and the solution epoch number at the 2nd 9th reference stations 51 The start time of data processing at the 1st kinematic station integer 4 kin_1_start_year integer 4 kin_1_start_month integer 4 kin_1_start_day integer 4 kin_1_start_hour integer 4 kin_1_start_min real 8 kin_1_start_sec real 8 kin_1_start_MJD 52 The specified epoch number at the 1st kinematic station integer 4 kin_1_num_epoch 53
27. ch controls the effect of outliers by the equivalent weight When there are anomalies both in the prior parameters and the measurement data it may lead to divergence of the solution In order to controlling the influences of the outliers the combination of outlier detection and robust estimation is adopted in HALO_GPS Firstly the large outliers are detected and moved out based on the median method in the data preprocessing Then the residual outliers are controlled by the robust estimation in the parameter adjustment The IGG3 scheme is applied to determine the equivalent weight The calculation formula of the equivalent weight is given below P yP 7 where P is the equivalent weight P is the original weight y is the adjusting factor of weight And the factor y is decided by the following formula 1 v lt k Ko k p l k lt y lt k 8 Y 5 bar o lt VSKk 8 0 v gt k 23 10 2312 GFZ b103 10119 Deutsches GeoForschungsZentrum GFZ where v is the standardized residual the value range of k is usually from 1 0 to 1 5 k is from 3 0 to 5 0 5 6 Adjustment Method There are numerous adjustment methods that can be used but least squares LS adjustment is the simplest and basic one For static case the LS adjustment algorithm is directly used for determining the complete unknowns For kinematic case there are two groups of unknowns One change with the time e g coordinates and another does not e g ambiguities A sequenti
28. e HALO_GPS Version 2 0 before the end of 2010 36 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 9 Acknowledgements Gratefully acknowledged are the supports from Prof Kahle Dr C F erste Dr F Barthelmes and Dr S Petrovic Without their encouragements this software would never been born I would like to thank my advisor Dr Guochang Xu for the guidance and encouragement given during my three years at GFZ I am also grateful to him for the freedom that he gave me to explore and develop my own ideas My special thanks go to my other co author Dr Tianhe Xu who is an excellent young scientist This software was significantly improved during the cooperation that we started last year He gave a key contribution for the development of the theory and algorithm I am also grateful to him for many discussions which made it possible the publications of some papers in peer reviewed journals I wish to have in the future many other occasions to cooperate with him In this study are used aircraft GPS data of the airborne gravity campaign NorthGrace2007 and AlpinAero2008 carried out by BKG in cooperation with GFZ and BGR the participating scientists and institutions are thanked for their cooperation Thanks Mr Markus Ramatschi and Dr Junping Chen of section 1 1 at GFZ for providing experimental equipments in antenna movement experiment Thanks Hong Kong Polytechnic University for
29. e IGS precise ephemeris to obtain the precise satellite clock corrections the precision is better than 0 1 nanoseconds Meanwhile the receiver clock error is estimated together with other parameters in station coordinate estimation In order to avoid the excess parameters a white noise model is usually used to express the behavior of receiver clock errors 5 3 Tropospheric Delay Correction A regional tropospheric model can be constructed using surveys from GPS ground networks Using this model the tropospheric delays of a kinematic station within the region can be interpolated However such a model is generally not suitable for an airborne platform high above the ground networks Therefore a method of constructing a regional tropospheric model for airborne GPS applications is developed and used in HALO_GPS First the kinematic station in the air is projected onto the ground Then the tropospheric delays at projected point are interpolated from those of the ground networks Finally the tropospheric delays at projected point are extended upward to the airborne platform using pressure and temperature gradients and humidity exponential function This method is called Projection Extension Method PEM for convenience of later discussion To investigate the impact of different tropospheric delays on GPS kinematic positioning some experiments were performed An IGS site site name OBE3 is taken as a fixed site a SAPOS site site name ZUGS
30. e observation The variation of position with time in a kinematic survey makes it hard to use a single observation for the detection of the cycle slips In a static GPS analysis it is possible to scan the pre processed residuals with a priori model for data quality control and to detect potential cycle slips at any epoch In a kinematic survey the pre fit residuals are of little use because of the movement of one receiver We use the two geometry free linear combinations of phase data the W M widelane L6 and the extra widelane L4 to detect potential cycle slips and outliers The calculate formula of L4 and L6 are as follows A A K 1 1 AN E 1 En ran f f BP L6 Q P Ge aN O N 2 ae rer ca E where 9 9 P P Ni N fis fa are the phase observations the pseudorange observations the ambiguities and the frequencies in L1 and L2 respectively is ionospheric delay and c is the speed of light The right hand side of Eq 1 shows that the residual contains only the L1 and L2 ambiguities and ionospheric delay Moreover the contribution of ionosphere is reduced by 65 by the factor 1 1 Zr fr of We re write Eq 1 as N LN 14 A s 3 A Af h Eq 3 denotes the frequency relationship directly and exclusively between the L1 and L2 ambiguity for each satellite from the phase observations If there were no cycle slips the temporal variations of the ionospheric residual in Eq 3 would be small for norm
31. e station 27 Ol l6th reference station 28 Ol 17th reference station 29 Ol 18th reference station 30 Ol I9th reference station 31 112008 09 25 08 07 44 0 Ilst reference station 33 1123226 Ilst reference station 34 Ol 2nd reference station 35 Ol 2nd reference station 36 Ol 3rd reference station 37 Ol 3rd reference station 38 Ol l4th reference station 39 Ol l4th reference station 40 Ol I5th reference station 41 Ol I5th reference station 42 Ol l6th reference station 43 Ol l6th reference station 44 Ol 17th reference station 45 Ol 17th reference station 46 Ol 18th reference station 47 Ol 18th reference station 48 Ol 9th reference station 49 Ol 9th reference station 50 112008 09 25 08 07 44 0 l1st kinematic station 51 1123226 Ilst kinematic station 52 17 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 1 Example result_file HALO_XYZ_1 txt laddress of result file 62 11 Example result_file HALO_BLH_1 txt laddress of result file 63 18 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 4 File Format 4 1 Input File Format A number of formats are currently used within the GNSS community for the exchange of data products and solutions The most important and widely accepted format is the RINEX Receiver INdependent EXchange format used for the exchange of GNSS observations broadcast information and meteorolog
32. es V B Langley R B Tropospheric zenith delay prediction accuracy for airborne GPS high precision positioning Proceedings of ION 54th Annual Meeting Denver Colorado June 1 3 pp 337 348 1998 Penna N Dodson A and Chen W Assessment of EGNOS Tropospheric Correction Model Journal of Navigation 54 1 37 55 2001 Saastamoinen J Contributions to the theory of atmospheric refraction Part II Refraction Corrections in Satellite Geodesy Bulletin Geodesique 107 13 34 1973 Sanssen V Ge L Rizos C Tropospheric delay corrections to differential INSAR results from GPS observations 6th International Symposium on SatNav 22 25 July Melbourne Australia 2003 Syndergaard S Retrieval analysis and methodologies in atmospheric limb sounding using the GNSS radio occultation technique Dissertation Niels Bohr Institute for Astronomy Physics and Geophysics Faculty of Science University of Copenhagen 1999 Schaefer U Liebsch G Schirmer U Meuschke A Pflug H Wang Q Petrovic S Meyer U AlpinAero2008 an airborne gravity campaign for improved geoid modelling in the Alps EGU 2009 Vienna Austria 19 24 April 2009 Troller M Geiger A Brockmann E et al Tomographic determination of the spatial distribution of water vapor using GPS observations Advance in Space Research 37 12 2211 2217 2006 Wanninger L Real time differential GPS error modeling in regional reference station networks Proc 1997 IAG Symposium 86 92
33. eutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 Introduction Niell global mapping function is adopted wet_mtt_map f90 Input parameters Temperature Latitude Height Elevation angle in deg of satellite Output parameters Wet mapping function Introduction Niell global mapping function is adopted spp_3 f90 Input parameters Kinematic station ID the structure variable IGS Output parameters XYZ and BLH coordinates of kinematic station kin_i_X epoch 1 kin_i_Y epoch 1 kin_i_Z epoch 1 kin_i_B epoch 1 kin_i_L epoch 1 kin_i_H epoch 1 The clock error of kinematic station kin_i_C epoch 1 Satellite elevation angle at each epoch kin_i_elev epoch sat_id The weight of observations kin_i_sat_p epoch sat_id Introduction This subroutine computes the initial positions of kinematic station and receiver clock error using Pseudorange C1 and the weight matrix of spp_1 solution again Additionally it deletes those satellites with the low elevation angle ref_clock_1 f90 Input parameters Reference station ID the structure variable IGS Output parameters The clock errors of reference station tef_i_C epoch 1 Satellite elevation angle at each epoch ref_i elev epoch sat_id The weight of observations ref_i_sat_p epoch sat_id Introduction This subroutine computes the clock error of reference station using Pseudorange C1 and known site coordinate ref_clock_2 f9
34. f satellite the structure variable IGS Output parameters XYZ and BLH coordinates of kinematic station kin_i_X epoch 1 kin_i_Y epoch 1 kin_i_Z epoch 1 kin_i_B epoch 1 kin_i_L epoch 1 kin_i_H epoch 1 The clock error of kinematic station kin_i_C epoch 1 Satellite elevation angle at each epoch kin_i_elev epoch sat_id The weight of observations kin_i_sat_p epoch sat_id Introduction This subroutine computes the initial positions of kinematic station and receiver clock errors using Pseudorange C1 based on the single point positioning solu_sat_xyzc f90 Input parameters The transmit time of signal Satellite ID Output parameters The position and clock error of satellite Introduction The subroutine uses 8 orders Chebyshev polynomial to interpolate satellite coordinates solu_ro_xyz f90 Input parameters Epoch number satellite ID station mark 1 reference station 2 kinematic station Output parameters The distance between satellite and reference station or kinematic station ref_i_RoxG sat ref_i_Roy j sat ref_i_Roz j sat ref_i_ROG sat kin_i_RoxG sat kin_i Roy j sat kin_i_Roz j sat kin_i_ROG sat Introduction This subroutine computes the distances between satellite and the station taking into account the influence of earth rotation xyz_blh f90 Input parameters XYZ Cartesian coordinates in WGS 84 system Output parameters BLH Geodetic coordinates in WGS 84 system Scie
35. he two sides of Eq 16 are multiplied by the matrix K where pele 19 0 8 E isaunit matrix Z M M 21 22 X 2 M M M M5M A P E A M A P A 21 where R U M M5U Al P E AMA P L 22 The Eq 20 can be divided into two parts M X R 23 M X M X U 24 Making J A M A P and considering E J E JXE J P E J E JYP then the Eq 21 and 22 can be expressed as M A P E J A A P E J E J A A E J P E J A 25 R A P E J L A E J PL 26 Assuming D E J A the Eq 23 can be written as D PD X D PL 27 The equivalent observation equation of Eq 27 is L DX V P 28 It is obvious that Eq 27 and Eq 13 are equivalent The solutions of them are identical The normal equation of Eq 14 can be written as X B PB X B PL 29 VAN a Since there are the same unknowns X X between two equations the Eq 27 can be added to the Eq 29 and then get 26 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 B PB x B PL 30 VAN 3 11 where B L are the design matrix and observation matrix after stacking respectively P is the weight matrix Finally the position parameters and ambiguity parameters of kinematic station can be estimated based on the Eq 30 27 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 6 Run of HALO_GPS T
36. her GPS Data 7 1 Mes 2 gt No If user chooses 1 the other data will be processed otherwise the program will exited 28 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 7 Numerical Examples For test run of the HALO_GPS a number of experiments have been performed In this Chapter some experimental results and analysis are given 7 1 Static Data Kinematic Processing In this experiment the GPS data of two fixed stations are used which are measured on February 1st 2010 at Hong Kong One site is taken as a reference station site name HKST the other site is treated as a kinematic station to estimate its coordinates at every epoch site name HKLT see Fig 4 0 5 10 Kilometers 4 A Satellite Positioning Reference Station Fig 4 Used Hong Kong GPS station network The sampling interval of observation data is 5 seconds 23 hours data were processed by HALO_GPS and the initial positions of HKLT are X 0 Y 0 Z 0 without any a priori information The positioning results are shown below 5389238 10 4 2399062 50 unit meters unit meters Y X 2399062 60 rn 2399062 70 539923720 uf N In M A 2399062 80 M smn 2399062 90 Mean 2399062 753 EE Mean 5389237 891 Std 0 018 1 Std 0 033 EELS LEE SEL SEES GLA GL LE DEL 5389237 60 E LAS SLA SL GLAS LA A O 1440 2880 4320 5760 7200 8640 10080 11520 12960 14400 15840 O 1440 2880
37. hnical Report No 203 University of New Brunswick Fredericton New Brunswick Canada 1999 Gendt G Reigber C Dick G Near real time water vapor estimation in a German GPS network First results from the ground program of the HGF GASP project Physics and Chemistry of the Earth A 26 6 8 413 416 2001 Gauthier L P Michel J Ventura Traveset and J Benedicto EGNOS The first step in Europe s contribution to the global navigation satellite system ESA Bulletin 105 35 42 2001 Hopfield H S Tropospheric effect of electromagnetically measured range Prediction form Surface Weather Data Radio Science 6 357 367 1971 Heise S Wickert J Beyerle G Schmidt T and Reigber Ch Global monitoring of tropospheric water vapor with GPS radio occultation aboard CHAMP Advance in Space Research 12 37 2222 2227 2006 Hu Guorong Ovstedal O Featherstone W E and et al Using the Virtual Reference Stations Concept for Long range Airborne GPS Kinematic Positioning J Survey Review 2008 40 307 83 91 King R W Bock Y Documentation for the GAMIT GPS analysis software version 10 03 Massachusetts Institute of Technology 2000 Mendes V B Collins P Langley R B The effect of tropospheric propagation delay errors in airborne GPS precision positioning Proceedings of ION GPS 95 the 8th International Technical Meeting of the Satellite Division of The Institute of Navigation Palm Springs Calif 12 15 September 1995 Mend
38. ical measurements The HALO_GPS Software Version 1 0 supports the following input file formats RINEX for the exchange of observation data broadcast information and meteorological data SP3 for the exchange of precise orbit and satellite clock information Troposphere SINEX Solution INdependent EXchange format for the export of troposphere information Clock RINEX for the exchange of satellite and receiver clock information 4 2 Output File Format For ease of use the products and solutions of this software the output file format of HALO_GPS is introduced below 1 The positioning results in Geodetic coordinates system 1 2 3 4 5 6 7 8 9 0 1 12345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890 2007 6 7 655 11 000 55 52214972 854791440 70438 0 002 0 002 0 004 0 005 2007 6 7 655 12 000 55 52214978 8 54791436 70 437 0 004 0 003 0 007 0 008 2007 6 7 655 13 000 55 52214976 8 54791445 70 448 0 006 0 004 0 010 0 013 2007 6 7 655 14 000 55 52214983 854791430 70433 0 016 0 012 0 030 0 036 2007 6 7 655 15 000 55 52214978 8 54791434 70 438 0 004 0 003 0 007 0 008 Year M Day H Min Second Latitude Longitude Height sigma B sigma L sigma H RMS Total 2 The positioning results in Cartesian coordinates system Ep Y M DHM Second x dX Y dY Z dz RMS DD 1 2010 1 2 0 0 0 00 2399062 732 0 010 5389237 877 0 026 2417327 081 0016 0032 7 2 2010 1 2 0 0 5 00 2399062 733 0 009 538923787
39. ingle I 2 array This routine is only valid for date after 1600 Jan 0 The relationship between JD and MJD is JD MJD 2400000 5d0 matinv f90 Input parameters The original matrix The rows of this matrix and the columns of this matrix Output parameters Inverse matrix svs_cm_to_phs f90 Input parameters MJD Satellite ID Satellite position of the center of mass Output parameters The phase center position of satellite Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 sun20 f90 Input parameters XMJD Epoch in modified Julian date in R 8 Barycentric Dynamical Time corresponding to ephemeris time Output parameters X K K 1 2 and 3 Rectangular coordinates of the sun in equatorial system J2000 0 in AU R Distance of earth sun in AU L B Introduction Ecliptical longitude latitude in mean system of epoch XMJD It is used to compute the position of the sun at XMJD time This subroutine was written using Simon Newcomb s tables of the sun read_ref_obsfile f90 Input parameters Address of observation file reference station ID start time the number of epoch Output parameters The height of antenna atthis station The number of observation types The specified types of observation The sampling interval GPS seconds at every epoch MJD at every epoch The number of satellites at every epoch All satellites ID at every epoch The number
40. ixes 2 List of Figures Figure 1 The data processing flowchart of HALO_GPS Figure 2 The residuals of tropospheric delays using RRM and PEM Figure 3 The residuals in height component using RRM and PEM Figure 4 Used Hong Kong GPS station network Figure 5 The results of static station kinematic processing Figure6 Used IGS reference station and kinematic station at GFZ Figure 7 The vertical motion experiment Figure 8 The positioning results on the height component Figure 9 The horizontal motion experiment Figure 10 The positioning results of motion distance on the horizontal component Figure 11 Sea buoy experiment at Repulse Bay Figure 12 The comparison of the positioning results between HALO_GPS and Ashtech Solutions Figure 13 The plane trajectory of 25 flights in NorthGrace2007 campaign Figure 14 The comparison of the positioning results between HALO_GPS and GAMIT Figure 15 Beech QueenAir 88 aircraft Figure 16 Location of mounted sensors on the aircraft Figure 17 The plane trajectory of 20 flights in AlpinAero2008 campaign Figure 18 The distance variation between two GPS antennas Figure 19 The comparison of the positioning results between HALO_GPS and GAMIT Figure 20 The comparison of the positioning results between HALO_GPS and TGO 41 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 42 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119
41. l hal E ih N Bar 2 5 30 5 20 Average 5 339 Std 0 013 0 3600 7200 10800 14400 Epoch seconds Fig 18 The distance variation between two GPS antennas September 26th 2008 October 13th 2008 nas HALO_GAMIT 04 HALO_GAMIT 024 0 2 E 5 E 07 o 3 en N A A ly 02 0 2 047 Mean 0 102 24 Mean 0 075 Std 0 036 Std 0 023 r r k r 1 r r i 1 0 3600 7200 10800 14400 18000 0 3600 7200 Epoch seconds Epoch seconds Fig 19 The comparison of the positioning results between HALO_GPS and GAMIT September 26th 2008 JOctober 13th 2008 oe HALO_TGO nen HALO_TGO 024 gt D 0 0 si Le A o gt nn eh dH meters dH meters Pr Mean 0 202 er Mean 0 177 g Std 0 033 Std 0 029 1 r T r T 0 3600 7200 10800 14400 18000 0 3600 7200 Epoch seconds Epoch seconds Fig 20 The comparison of the positioning results between HALO_GPS and TGO 35 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 8 Summary HALO_GPS is developed at GFZ to achieve cm level accuracy for an aircraft trajectory for application in airborne gravimetry Some new strategies and algorithms are adopted to deal with complex environmental conditions in aircraft positioning such as robust estimation median method and fast ambiguity resolution Furthermore we developed the methods of automatically choosing and changing reference satellite and the
42. n to the number of coefficients receiver_antenna_correction f90 Input parameters Longitude of station Latitude of station Height of antenna at this station ref_i_Hant ref_i_Eant ref_i_Nant Output parameters The corrections of receiver antenna phase centre ref_i_ant_x ref_i_ant_y ref_i_ant_z kin_i_ant_x kin_i_ant_y kin_i_ant_z unify_obs_time f90 Input parameters Reference station ID kinematic station ID Output parameters The real number of processing epochs of reference station and kinematic station Introduction Firstly this subroutine tries to find out the lost epochs in reference station and kinematic station Then it removes these epochs from the observations of reference station which are lost in that of kinematic station and removes these epochs from the observations of kinematic station which are lost in that of reference station Finally the epochs between reference and kinematic stations are arranged in a certain order mjd_to_ymdhms f90 Input parameters Modified Julian Date MJD Output parameters Year Month Day Hour Minute Second Note If a full Julian date is used the resolution of the seconds will only be about 10 microseconds and a MJD should yield a resolution of about 0 1 microseconds in the seconds Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 spp_1 f90 Input parameters Kinematic station ID position and clock error o
43. ncononanonncononn A conc cnnonnc as 29 7 2 Antenna Movement Experiment h s onien n i e i e e 30 3 Sea BUOY Experiment ohni e a a sans 32 7 4 NorthGrace2007 Campaign ccc paii o aai a A i in 32 73 Alpin Acro2008 Campaign 2282 RE ld bee pn 33 A NN 36 O Y AA banpetescdviedussunaedeordonedus seubelnosdvasleplavsesuenashslescevunwbednesss 37 10 References iaa neds tise dus ER Ei an dea ne acia cda 38 TAA PPENGI KES oe Br Al ver aseastuhvontents 40 11 1 Appendixes 1 Definitions of Constants uesersesnennensessnennennennonnnennennennannnennennennan 40 11 2 Appendixes 23 List of Figures asocion n namens aa aai 41 2 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 1 Introduction HALO_GPS is a precise GPS kinematic positioning software It was developed at GFZ Potsdam for the German HALO project The goal is to develop a software which is able to achieve cm level accuracy for an aircraft trajectory for application in airborne gravimetry To fulfill the needs of the HALO project some new algorithms and strategies are developed and adopted in this software It can automatically choose and change reference satellite as well as the reference station automatically detect cycle slips outliers bad observation data and potential large jumps in the receiver clock The one click functionality is implemented for ease of use All process steps will be finished automatically after the
44. ntific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 spp_2 f90 Input parameters Kinematic station ID the structure variable IGS Output parameters The tropospheric delay of kinematic station at every epoch kin_i_trop epoch sat_id Introduction kin_i_tropG sat_id kin_i_dry_ztd j 1 kin_i_dry_mf j sat_id kin_i_wet_ztd j 1 kin_i_wet_mf j sat_id met_seasonal f90 Input parameters JD Latitude Height Output parameters Temperature in Celsius Pressure in mbar Relative humidity Bias in surface temperature Introduction This subroutine computes the temperature pressure and relative humidity based on the seasonal argument dry_saas_zen f90 Input parameters Temperature Latitude Height Pressure Output parameters Dry zenith delay of the station Introduction Routine to compute dry zenith delay based on Saastamoinen model wet_saas_zen f90 Input parameters Temperature Latitude Height Relative humidity Output parameters Wet zenith delay of the station Introduction Routine to compute wet zenith delay based on Saastamoinen model wet_press f90 Input parameters Temperature Relative humidity Output parameters The partial pressure of water vapor dry_mtt_map f90 Input parameters Temperature Latitude Height Elevation angle in deg of satellite Output parameters Dry mapping function Scientific Technical Report STR 10 11 D
45. o start the HALO_GPS the user just needs to enter the following command lines HALO Then the following information will be shown on the user screen HALO_GFS High Altitude and Long Range Airborne GPS Positioning Software Created Date March 2614 Created By Qianxin Wang Advisors Dr Tianhe Au Dr Guochang Xu Email kingsen gfz potsdan de Telephone 8649 331 2838 1187 Copyright Helmholtz Centre Potsdam German Research Centre for Geosciences Version 1 8 A Reading Control File Please Choose 1 gt Run Example 2 gt Run Special Control File 3 gt Exit To testing the installation of HALO_GPS software we provide three examples to user If the user enters 1 the program will run the examples automatically If the user chooses 2 the address of special control file will be asked to input The control file consist all necessary information and control parameters for running this software And the user control file has to be edited before running the program The format and definitions of user s own control file have been introduced in Chapter 3 The whole data processing is automatic After all data is processed successfully the program will give the following information Sat 23 Period dd_amb_LG Epoch 1 28 522 2598 3608 Sat 28 Period dd_amb_LG Epoch 1 8 763 1 3668 Sat 32 Period dd_amb_LG Epoch 1 51 961 1 2218 Successful Solutiont Do You Want to Process Ot
46. ontrol file for an aircraft kinematic positioning Explanations will be outlined after this input parameter file The control file of all other numerical tests given in this manual can be obtained through minor modification from this standard input control file 15 Scientific Technical Report STR 10 11 10 2312 GFZ b103 10119 Deutsches GeoForschungsZentrum GFZ User Control File Created by Qianxin Wang February 2010 Technical Advisor Tianhe Xu Guochang Xu 1 Example data_file AlpinAero2008 igs 14984 sp3 ISp3 file 1 Ol Broadcast Ephemeris 2 11 Example data_file AlpinAero2008 opaf2690 080 Ilst reference station 3 Ol 2nd reference station 4 Ol 3rd reference station 5 Ol l4th reference station 6 Ol I5th reference station 7 Ol l6th reference station 8 Ol 17th reference station 9 Ol 18th reference station 10 Ol l9th reference station 11 1IOPAF Ilst reference station 13 Ol 2nd reference station 14 Ol 3rd reference station 15 Ol 14th reference station 16 Ol 5th reference station 17 Ol loth reference station 18 Ol 17th reference station 19 Ol I8th reference station 20 Ol 10th reference station 21 6 Reference Station Coordinate 16 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 114186557 029 1835026 434 14723761 505 Ilst reference station 23 Ol 2nd reference station 24 Ol 3rd reference station 25 Ol l4th reference station 26 Ol I5th referenc
47. ore as a fixed station and another receiver is installed on a buoy in the sea see Fig 11 Firstly we estimate the position of GPS buoy using a well known GPS commercial software Ashtech Solutions 2 60 Since the reference station is very close to the GPS buoy about 150 meters most GPS measurement errors can be cancelled out by the relative position mode Therefore the position accuracy of centimeter level can be easily achieved for this experiment Fig 12 is a comparison of the positioning results between HALO_GPS and Ashtech Solutions on the horizontal and height components Fig 11 Sea buoy experiment at Repulse Bay December 8th 2004 December 8th 2004 04 HALO Ashtech a HALO Ashtech dH meters o dE meters Mean 0 033 Mean 0 030 aac Std 0 021 j Std 0 027 r T T T 3600 7200 o 3600 7200 Epoch seconds Epoch seconds Fig 12 The comparison of the positioning results between HALO_GPS and Ashtech Solutions Fig 12 shows the differences between HALO_GPS and Ashtech Solution are very flat The means and standard deviations are 2 3 cm 7 4 NorthGrace2007 Campaign Since HALO_GPS is developed to fulfill the need of precise positioning in airborne gravimetry it has to be rigorously tested with real aircraft GPS data We used it to process the GPS data of 32 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 NorthGrace2007 and AlpinAero2008 airborne
48. positioning precision in height component of PEM is 8 mm better than that of RRM In some applications e g airborne gravity determination it is just the height component which is of ultimate importance 5 4 Ambiguity Resolution Unlike other factors which have common sources the phase bias is receiver channel dependent In GPS receiver each channel initializes its own counter for one satellite so the phase biases can not be canceled by differencing and modeling Resolving unknown phase bias becomes a fundamental requirement for accurate cm level GPS aircraft kinematic positioning Moreover ambiguity resolution is more crucial in the kinematic surveying than in the static one In static GPS analysis we can separate the ambiguities from the receiver s fixed position by the geometric changes of satellite in a long time observation In kinematic GPS surveying the occupation of receiver in one location is short even varies from epoch by epoch The ambiguities are high correlated with positions Therefore we feel the strong need to develop an ambiguity resolution method for aircraft positioning to be not only fast but also reliable having the capability to deal with complex environmental conditions There is not a simple way to develop a comprehensive approach for ambiguity solution considering the complex field conditions in which GPS surveys performed In HALO_GPS software an ambiguity search in the ambiguity space is done and include
49. providing the data of sea buoys And thanks Massachusetts Institute of Technology for providing GAMIT software Thanks my colleagues Indridi Einarsson Roelof Rietbroek and Dr Magdala Tesauro for their help revise this manual to make it more readable This work was sponsored by GFZ PhD Student Scholarship China Scholarship Council and the Helmholtz Association of German Research Centers Council 37 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 10 References Boehm J Niell A Tregoning P Schuh H Global Mapping Function GMF A new empirical mapping function based on numerical weather model data Geophysical Research Letter 33 L07304 2006 Chen Gang GPS Kinematic Positioning for the Airborne Laser Altimetry at Long Valley California D PhD thesis 1998 Massachusetts Institute of Technology U S A Christoph Foeste Mirko Scheinert A Platform for Earth Observations and Geophysics R DFG Priority Program 1294 HALO Evaluation Colloquium Oberpaffenhofen Germany 11 12 March 2010 Collins J P Langley R B Estimating the residual tropospheric delay for airborne differential GPS positioning Proceedings of ION GPS 97 1197 1206 Kansas City Mo 16 19 September 1997 Collins J P Assessment and development of a tropospheric delay model for aircraft rovers of the global positioning System M Sc E Thesis Department of Geodesy and Geomatics Engineering Tec
50. reference station to fulfill the needs of HALO project This software has been tested with many kinds of real data Comparisons have been made with several well known GPS software packages The results show the kinematic positioning accuracy of HALO_GPS is about 2 cm 5 cm Of course any GPS software can not obtain a satisfying result when the data quality is too bad in very few epochs Through processing a number of real data from NorthGrace2007 and AlpinAero2008 campaign the stability and reliability of this software are validated For the beginner the one click functionality is implemented for ease of use All process steps will be finished automatically after the user enters one command The application programming interface API is also provided for the professional users to develop their own functions The source code of HALO_GPS software is opened for the researcher to study and communication Additionally for testing the installation of HALO_GPS software some examples and standard control file templates are prepared for the user Although HALO_GPS Version 1 0 is well qualified to process the standard aircraft GPS data and has strong stable reliable as well as high precision it is not a strong function software until now However some functions e g filter algorithm network solution precise single point positioning and GPS INS integrated positioning have been finished If possible these functions will be implemented in th
51. s the following five steps 1 Selection of an initial search center 2 Selection of a search space 22 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 Scientific Technical Report STR 10 11 3 Reduction of search candidates by the constraint conditions of geometry and physics 4 Searching for the best candidates 5 Significance check and verification The initial ambiguities are obtained by the following equations _ VAN 4x4 VAN 6 XA VAN 4 Ll A A V V VAN u VN A ae lt 5 2 VAN fi VAN Diy VAN 6 LC f 9 Il f L2 1 2 1 where VAN VAN is the double differenced ambiguities of the extra widelane L4 and the W M widelane L6 The VAN VAN can be obtained by the Eq 2 3 The detailed introduction of this method is given in the references 5 5 Robust Estimation Since the precision of GPS phase observations are better than 1 mm it is usually used for the precise positioning However there inevitably exist observational outliers in the real surveying If the outliers are not eliminated or controlled the estimated parameters will be distorted There are two commonly used methods to control the outliers One is the outlier detection which detects the outliers by statistic test Since this method is usually based on the least squares LS adjustment the statistic will be influenced by the outliers The other is robust estimation whi
52. sing a Beech QueenAir 88 aircraft see Fig 15 Fig 17 shows the plane trajectory of 20 flights in this campaign For geodetic positioning of the aircraft one Novatel OEM 4 and one Topcon NET G3 GPS receivers together with a GPS controlled inertial measurement unit Aerocontrol IIb of IGI Company were used Data sampling rate was 10 Hz The two GPS antennas were located at the nose and near to the tail of the aircraft respectively see Fig 16 The Euclidian distance between the antennas was 5 343 meters For evolution of HALO_GPS a number of internal and external comparisons are performed Fig 18 shows the distance variation between two GPS antennas on October 13th 2008 The distance is obtained by calculating the separate positioning results of two antennas without any a priori information Fig 19 is a comparison of the positioning results between HALO_GPS and GAMIT software on the height component The flights are on September 26th 2008 and October 13th 2008 respectively Fig 20 is a comparison between HALO_GPS and commercial software Trimble Geomatics Office TGO at the same days Comparisons show good performance of HALO_GPS The standard deviation is better than 5 cm GPS antenna Gravimeier Fig 17 The plane trajectory of 20 flights in AlpinAero2008 campaign 34 Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 5 60 October 13th 2008 5 50 E 5 40 E Al Al ll l
53. user enters one command The development of the HALO_GPS software was started in 2009 first year for theoretical study and then for code design It has been tested with various kinds of real data Many comparisons have been made with several well known GPS software packages such as Ashtech Solution Trimble Geomatics Office and GAMIT The results show a strong stability and reliability of HALO_GPS The software has been used to successfully process the GPS data of NorthGrace2007 and AlpinAero2008 airborne gravimetry campaigns This manual outlines the characteristics of the software and describes how to use it The principles and new features are outlined systematically and referred partly to existing references The major functions of some important subroutines are introduced briefly Numerical examples of kinematic positioning and internal tests as well as external comparisons are given This software is developed in Fortran 90 under Unix operating system and can be used on PCs under Linux without any change The user interface to HALO_GPS is command driven with default values for most processing This interface provides flexibility and should make the software usable with little training Scientific Technical Report STR 10 11 Deutsches GeoForschungsZentrum GFZ 10 2312 GFZ b103 10119 2 Structure of Software HALO_GPS software consists of a main function and about 50 important subroutines Each subroutine attends to its own duties which

Download Pdf Manuals

Related Search

Related Contents

Eglo 20956A Installation Guide  HP V 1905-8-PoE  PCS 91.B&R - Van Egmond Groep  ViewSonic OPTIQUEST Q2205WB User's Manual  Encofrado Horizontal CC-4(INGLES) OK.qxp - Uni-span  Resistance Decade Box - GME Technology, LLC.  Samsung Reproductor de Blu-ray E5300 Manual de Usuario  

Copyright © All rights reserved. Failed to retrieve file