α 1. Φ2 = MPp 1 + B 1 + MΦ 2. α 1. α 1. MP1 is then defined as the linear combination obtained (eqn 8): α 1. α 1

Transcription

1 The phase range equation (eqn 5) is a linear combination of the L1 and L2 phase observables that gives us the range to the satellite along with receiver/satellite clock errors, tropospheric errors, a bias term and a phase multipath term. Combining the equation for ionospheric delay (eqn 4), the phase range equation (eqn 5) and the pseudo-range equation (eqn 1) gives us: p (eqn 8) α Φ α 1 Φ2 = MPp 1 + B 1 + MΦ 1 Where B 1 is the bias term defined as: B1 ( λ1 N1 λ2 N2 ) b (eqn 9) α = = 1 λ1 N α 1 λ2 N 2 and MΦ 1 is the phase multi-path term defined as: MΦ 1 ( MPΦ 1 MPΦ 2 ) mφ (eqn 10) α = = MPΦ α 1 MPΦ2 MP1 is then defined as the linear combination obtained (eqn 8): MP1 p α 1 2 Φ α 1 Φ2 Performing similar operation for L2, we get the following equations: p 2α α 1 2α Φ α 1 Φ2 = MPp 2 + B 2 + MΦ 2 (eqn 11) (eqn 12) B 2 = 2α α 1 λ1 N 2α α 1 λ2 N 2 (eqn 13) MΦ 2 = 2α α 1 2α MPΦ α 1 MPΦ2 (eqn 14) MP2 p 2α α 1 2α Φ α 1 Φ2 (eqn 15) Looking at multipath equations for L1 (eqn 8) and L2 (eqn 12), we see that the MP equations contain two terms in addition to the multipath values. The first term is a bias term ( B i ) that arrises from the unknown phase ambiguities. QC fixes this value to be the first MP value calculated for a satellite. This is not strictly true, but brings the MP values to near zero values. What is of interest from the MP values is it s structure over time, not the DC bias term. The second addition term ( MΦ 1 ) comes from the phase multi-path. Since the phase multipath is much smaller in magnitude than the P-code multipath, the MP values are dominated by P-code multipath. Page 30 Updated 3/7/95

2 Derivation of MP1 and MP2: First we describe the pseudorange measurements as: p i = R + c( dt dt) + I i + I + MPp i (eqn 1) and the phase measurements as: Φ i = R + c( dt dt) + λ i N i I i + I + MPΦ i (eqn 2) Where: p R c dt = pseudo-range observation in meters = distance between satellite and receiver in meters = speed of light in meters/sec = satellite clock error in sec = receiver clock error in sec dt I = ionospheric range error in meters I = tropospheric range error in meters N = integer cycle ambiguity MPp = pseudo-range multipath MPΦ = phase multipath i = observation frequency, i = 1 L1; i = 2 L2 f 1 = Frequency of L1; f 1 = GHz f 2 = Frequency of L2; f 2 = GHz λ 1 = Wavelength of L1; λ cm λ 2 = Wavelength of L1; λ cm By taking advantage of the fact that the ionospheric delay for L1 and L2 are related as: f 2 1 I 2 = αi 1 ; α --- (eqn 3) f 2 We can solve the phase measurement equation for the ionospheric delay. First we will solve for I 1 : ( Φ 1 Φ 2 ) (eqn 4) ( α 1) I ( λ N λ N ) ( MPΦ MPΦ 2 ) = + 1 ( α 1) ( α 1) This equation combined with the phase equation for L1 (eqn 2) gives us the phase range equation: ( Φ Φ 1 Φ 2 ) = R + c( dt dt) + I+ b (eqn 5) ( α 1) 1 + mφ 1 Where B 1 is the phase bias term defined as: ( λ 1 N 1 λ 2 N 2 ) b1 = λ1 N (eqn 6) ( α 1) and MΦ 1 is the phase multi-path term defined as: ( MPΦ 1 MPΦ 2 ) mφ1 = MPΦ (eqn 7) ( α 1) Page 29 Updated 3/7/95

3 Compact Format: On of the output formats for plot files in QC is the COMPACT format. This section describes this format. COMACT plot files are plain ASCII files which contains a number of header lines followed by data lines. Below is the beginning of a COMPACT file with annotations: The first line indicates that the file is in COMPACT format COMPACT The second line lists the SV s that are contained in this file. The compact format is able to handle either single SV s are SV pairs. If data is recorded only for single SV s, the SV number is repeated. To following line is really a single line in the computer, although it is wrapped in this document. SVS 02:02 07:07 15:15 27:27 19:19 26:26 13:13 04:04 09:09 12:12 24:24 05:05 20:20 16:16 03:03 17:17 23:23 22:22 21:21 28:28 01:01 31:31 25:25 14:14 29:29 18:18 Next comes the sampling rate of the data file in seconds. T_SAMP 30.0 This is the time of the first epoch in modified julian data. START_TIME_MJL The data starts here. Normally there are two lines for each data record, unless there is no data for any SVs, as in this case. The 0 here means that for this data point, there is data for zero satellites 0 This data line is a complete pair. The first line has the number of SV s there is data for, followed by an index to the SVS header line for each SV. If the number of SV s is -1, then the previous number of SVs and SV indices are used. The second line of the pair contains the actual data. In this example, the first data point is for the first SV listed in the SVS header line (02 in this case) Here we have data for the same SVs in the same order, and thus we have only a -1 to indicate that there is no change in the SVs Page 28 Updated 3/7/95

4 QC Plot Files: If the appropriate options are set in the qc.inp file, then QC will generate plot output files. The names of these plot files will have the same base name as the RINEX observation file. For example, for the RINEX file math o, plot files would have the names math0770.ext where ext is on of the following: Table 3: Types of plot files QC generates Extension.mp1.mp2.ion.iod.azi.ele Type of plot file Values for L1 code multipath. Values for L2 code multipath. The ionospheric delay. The first derivative of the ionospheric delay. Azimuth angles to the SV s. Elevation angles to the SV s. Page 27 Updated 3/7/95

5 QC Symbol File (qc.sym): The file qc.sym controls which symbols are used in the QC summary character based plot. If no qc.sym file is present in the directory where QC is started, then standard default values are used. Below is an example of a qc.sym file with the standard default characters specified for each flag:. - L1 C/A only no A/S, - L1 C/A only A/S : - L1 P only no A/S ; - L1 P only A/S ~ - L1 C/A L2 P no A/S # - L1 C/A L2 P A/S * - L1 P L2 P no A/S Y - L1 P L2 P A/S Page 26 Updated 3/7/95

6 Meaning of flags: I slip detected on iono phase S multipath slip MP1 and MP2 R multipath slip on MP1 only P multipath slip on MP2 only C clock reset / slip (optional) G gap in data - SV up but no data found + SV data but below elev mask. L1 C/A only no A/S, L1 C/A only A/S : L1 P only no A/S ; L1 P only A/S ~ L1 C/A L2 P no A/S # L1 C/A L2 P A/S * L1 P L2 P no A/S Y L1 P L2 P A/S RINEX HEADER: 2 OBSERVATION DATA G (GPS) RINEX VERSION / TYPE ASRINEXO V2.5.4 LH SOPAC/IGPP/SIO/UCSD 20-FEB-95 04:00 PGM / RUN BY / DATE COMMENT PGGA Data [Permanent GPS Geodetic Array - S. California] COMMENT COMMENT SOPAC (Scripps Orbit and Permanent Array) COMMENT IGPP [Institute of Geophysics and Planetary Physics] COMMENT SIO [Scripps Institution of Oceanography] COMMENT UCSD [University of California, San Diego] COMMENT COMMENT questions: pgga@pgga.ucsd.edu COMMENT GPS data archive anonymous login: toba.ucsd.edu COMMENT BLYT JAB MARKER NAME OBSERVER / AGENCY ASHTECH Z-XII3 1D00 REC # / TYPE / VERS 941 GEODETIC L1/L2 P ANT # / TYPE APPROX POSITION XYZ ANTENNA: DELTA H/E/N 1 1 WAVELENGTH FACT L1/2 5 C1 L1 L2 P1 P2 # / TYPES OF OBSERV 30 INTERVAL TIME OF FIRST OBS END OF HEADER END OF HEADER Notice that the presence of data for SVs with A/S on are reported with the Y character instead of the # in the TurboRogue and Trimble examples. This is because the Ashtech Z-XII RINEX files report a P1 and P2 code value under A/S. TurboRouge and Trimble receivers report C/A for the P1-Code. Page 25 Updated 3/7/95

7 Now we have an example summary file from an Ashtech Z-XII receiver. The same input options for BLYT S were used to generate this file. QC v3 by Summary File: BLYT S Receiver type: ashtech z-xii S 1 YYYYYYYYYYYYYYYYYY++ A 2 YYYYYYYYY+ +IYYYIYYYYYYY T 4 YYYYYYYYYYYYYYY++ E 5 +YYYYYYYYYYYYYYYY+ L 6 +YYYYYYYYYYYYYYYYYY++ L 7 YYYYYYYYYYY++ IYYYYYY I 9 +YYYYYYYYYYYYY++ +IYYYYY++ T 12 ***************++ I E 14 +YYYYYYYYYYYYYYYY+ 15 Y++ IIYYYYYYYYYYY++ ++IYYY 16 +YYYYYYYYYYYYYY+ +IYY YYYYYYYYYYYYYYYYYYYY YYYYYYYYYYYYYYYYY IYYYYYYYYYYYYYYYYYYY 20 +YYYYYYYYYYYYYY YYYYYYYYYYYYYYYYYYYY YYYYYY++ IYYYYYYYYYYYY YYYYYYYYYYYYYYYYYYY YYYYYYYYYYYYYYYYY+ 25 +YYYYYYYYYYYYYYY++ 26 YYYYY++ +SYYYYYYYYYY++ +SYYY 27 YYYY++ +SYYYYYYYYYYYYYY 28 +YYYYYYYYYYYY+ +IYYYYYYY YYYYYYYYYYYYYYYYYYY++ 31 YYYYYYYYY++ IYYYYYYYYYYY+ CLK CC CC C CC CC CC CC CCC CCCCC CCCCCCC CCCCCCCCCCCCCC CCCCC CCC CC CC CC :00 23:59 Time of First Epoch in File (year,month,day,hour) : : 0 Time of Last Epoch in File (year,month,day,hour) : :59 Observation Interval for File (in seconds) : 30 Elevation cutoff for qc : 15 Total number of observations expected : Total number of observations in file : Total number of points deleted : 2214 Data collection percentage : 100 RINEX vs qc point pos diff [Km] : 0.05 Average MP1 : Average MP2 : # of points for MP moving average : 50 Average clock drift [msec/hr] : Average time between resets [min] : Number of detected slips : 17 Observations per slip : 1062 first epoch last epoch hrs dt #expt #have % mp1 mp2 o/slp SUM : : Page 24 Updated 3/7/95

8 Meaning of flags: I slip detected on iono phase S multipath slip MP1 and MP2 R multipath slip on MP1 only P multipath slip on MP2 only C clock reset / slip (optional) G gap in data - SV up but no data found + SV data but below elev mask. L1 C/A only no A/S, L1 C/A only A/S : L1 P only no A/S ; L1 P only A/S ~ L1 C/A L2 P no A/S # L1 C/A L2 P A/S * L1 P L2 P no A/S Y L1 P L2 P A/S RINEX HEADER: 2 OBSERVATION DATA G (GPS) RINEX VERSION / TYPE TRRINEXO V2.4.6 LH L+T 20-FEB-95 00:41 PGM / RUN BY / DATE Zimmerwald LT88 COMMENT BIT 2 OF LLI (+4) FLAGS DATA COLLECTED UNDER AS CONDITION COMMENT ZIMM MARKER NAME LOGST/COMPAQ L+T OBSERVER / AGENCY 2691 TRIMBLE 4000SSE 6.10 REC # / TYPE / VERS ST L1/L2 GEOD ANT # / TYPE APPROX POSITION XYZ ANTENNA: DELTA H/E/N 1 1 WAVELENGTH FACT L1/2 5 C1 L1 L2 P2 P1 # / TYPES OF OBSERV 30 INTERVAL TIME OF FIRST OBS END OF HEADER END OF HEADER Notice that this file does have CLK events. Page 23 Updated 3/7/95

9 Next we have an example summary file from a Trimble-SSE receiver. The same input options for ALBH S were used to generate this file. QC v3 by Summary File: ZIMM S Receiver type: trimble 4000sse S 1 #######++ I##########R+ +I## A 2 #####++ S#############++ + T 4 +#############++ +I####++ E 5 ##################++ L 6 +################I+ L 7 ##########RI+ +II########+ I 9 ##################++ T 12 +******************++ E 14 ##################++ 15 #############++ +I### 16 ############### ############## #################++ 19 ############### I#################++ 21 ###++ +I#I#I+ +I########## ############### I#######++ II############ 24 ############++ +I######## ######++ +I############+ 26 +################ ################++ 28 ###+ I###I########## 29 +################++ 31 ######++ I######### CLK CC CC C CC CC CC CC CC C CC CC CC CC CC C CC CC CC CC CC CC C CC CC CC :00 23:59 Time of First Epoch in File (year,month,day,hour) : : 0 Time of Last Epoch in File (year,month,day,hour) : :59 Observation Interval for File (in seconds) : 30 Elevation cutoff for qc : 15 Total number of observations expected : Total number of observations in file : Total number of points deleted : 2376 Data collection percentage : 99 RINEX vs qc point pos diff [Km] : 0.17 Average MP1 : Average MP2 : # of points for MP moving average : 50 Average clock drift [msec/hr] : Average time between resets [min] : Number of detected slips : 27 Observations per slip : 647 first epoch last epoch hrs dt #expt #have % mp1 mp2 o/slp SUM : : Page 22 Updated 3/7/95

10 Below the character based graph of data contained in the RINEX file is a summary of other information. This information is also displayed in the output that QC generates and is repeated here for conveinance. In most cases looking at the summary file should be sufficient to determine if there are problems with a RINEX file. If problems are found, then the user can look at the more detailed output QC generates to further investigate the problem. Notice that in the above example the Average clock drift [msec/hr] and Average time between resets [min] values are both zero. This is because the data came from an Allen-Osborn Turbo-Rouge receiver and there are no clock resets in this instrument type. For these types of receivers both of these values are set to zero. After these data values, there is a single line that starts with the word SUM that has the most pertinent information for the summary file. Above the line is a header line with short descriptions of the data values. The purpose of this line is to allow the user to extract the most important information from the summary file by simply searching for a line that starts with the word SUM (i.e. grep ^SUM ALBH S from UNIX would extract this summary line). The first two values in this line are the start and end times of the first and last epochs, respectively (first epoch and last epoch), followed by the length of the data set in hours (hrs), the sampling interval in seconds (dt), the number of data points QC expected to find in the RINEX file (#expt), the number of data points found in the data file (#have -- data points below the elevation cut of angle specified in the qc.inp file are not counted), the data collection percentage (% -- calculated as #have/#expt * 100), the average of the MP1 RMS values for all SV s (mp1), the average of the MP2 RMS values for all SV s (mp2) and finally the ration of observations/slips (o/slp -- if there are no slips in the file, then the total number of observations are reported here). Next comes a description of the flags used in the character based plot of data contained in the RINEX file followed by (if requested in the qc.inp file) the contents of the header in the RINEX file (lines between RINEX HEADER: and END OF HEADER). Page 21 Updated 3/7/95

11 It is important to know that for this graph, time has been divided up in to 63 bins and that more than one observation will map to the same bin. This means that if there is at least one data point for a SV in the time bin, then QC will indicate that there is data for the entire bin (which is a character position). The last line of the plot is labeled CLK. Here QC will display a C when it finds a 1 ms jump in the time that GPS data is collected. The tick marks on the X-axis are in intervals of three hours. In the above example, there are no CLK events. This is because the data is from an TurboRogue, which steers its internal clock. At the end of this section we will show examples for a Trimble-SSE and an Ashtech Z-XII, which will have CLK events. Besides the presence and absence of data, this plot will indicate other aspects of the data with different flags. The meaning of these flags are listed below (they are also output at the end of the.yrs file). QC will also display any slips it finds in the data. Below is a summary of the different slip flags used by QC and their meanings:: Table 2: QC Flag Summary Flag R P B S I C G Meaning This flag indicates that there was a change in the calculated MP1 value high enough to indicate that the MP1 values were high. Same as the R flag except for MP2. This indicates that both an R and P flag occurred at the same epoch. This flag indicates that there was a jump in the MP1 and MP2 values high enough to indicate that there was a slip in either the L1 and or L2 phases. This usually occurs when both the L1 and L2 phase values get reset to zero after loss of lock. Since both L1 and L2 change, a slip does not show up in the ionospheric value. This flag indicates that there was a change in the computed ionospheric delay to indicate that there was a slip in L1 and/or L2. Clock reset or clock slip flag. In the bottom line of the graph (the one labeled CL), a C indicates that there was a 1 msec change in the time GPS data was collected. If a C occurs in a line for a satellite, it indicates that there was a clock slip -- which means that there was a 1 msec change for all sv s at the same epoch. The flag indicates that there was a data gap in the RINEX file at this time. The G flag will appear for all SV s regardless of whether or not they were being tracked before or after the gap. In addition, the G s will be located at the end of the gap. NOTE: The A/S flags are determined by looking at the data types and flags contained in the RINEX file. If the flags in the RINEX file are not correct, then the graph will not correctly show the times for which A/S was on. Page 20 Updated 3/7/95

12 first epoch last epoch hrs dt #expt #have % mp1 mp2 o/slp SUM : : Meaning of flags: I slip detected on iono phase S multipath slip MP1 and MP2 R multipath slip on MP1 only P multipath slip on MP2 only C clock reset / slip (optional) G gap in data - SV up but no data found + SV data but below elev mask. L1 C/A only no A/S, L1 C/A only A/S : L1 P only no A/S ; L1 P only A/S ~ L1 C/A L2 P no A/S # L1 C/A L2 P A/S * L1 P L2 P no A/S Y L1 P L2 P A/S RINEX HEADER: 2 OBSERVATION DATA G (GPS) RINEX VERSION / TYPE RGRINEXN V2.1.0 VM EMR 20-FEB-95 05:00 PGM / RUN BY / DATE BIT 2 OF LLI (+4) FLAGS DATA COLLECTED UNDER AS CONDITION COMMENT HARDWARE CALIBRATION (S) COMMENT STATION INFORMATION UPDATED COMMENT ANT HGT.110 M BELOW L1 AND.128 M BELOW L2 PHASE CENTRE COMMENT albh WCDA-ACP VICTORIA, B.C., CANADA MARKER NAME AUTO-DOWNLOAD NATURAL RESOURCES CA OBSERVER / AGENCY 292 ROGUE SNR-8000 TURBO REC # / TYPE / VERS 368 DORNE MARGOLIN T ANT # / TYPE APPROX POSITION XYZ ANTENNA: DELTA H/E/N 1 1 WAVELENGTH FACT L1/2 5 P1 L1 L2 P2 C1 # / TYPES OF OBSERV 30 INTERVAL TIME OF FIRST OBS END OF HEADER END OF HEADER (NOTE: Text in the right most column of the RINEX HEADER: section were edited to fit on this page. The data description comment should all start in the same column as the COMMENT lines.) At the top of the output is a graphical display of the data contained in the RINEX observation file. On the X-axis time is displayed along with the time of day for the first and last epochs. Each vertical bar (tick mark) is separated by 3 hours. On the Y-axis the SV ids of the satellites tracked appear. Along each row the presence or absence of data is displayed, along with flags that indicate which frequencies code data was available for, whether or not A/S was on and in the case of L1, if the pseudo-range code was P-code or C/A-code. The meaning of each flag is listed near the end of the summary file and the actual character symbols used can be changed by setting new values in the qc.sym file. The character symbols shown here are the default values (the ones that would be used if no qc.sym file were present). In addition to showing when data exists for each satellite, QC indicates when there is missing data for a satellite (indicated by a - ) and when there is data for a satellite when it is below the elevation cut off specified in the qc.inp file (indicated by a + sign). Page 19 Updated 3/7/95

13 QC Summary File (.yrs): QC will generate an ASCII summary file that will have the same name as the RINEX observation file except that the last character will be an S. Below is an example of a summary file generated by QC (this output came from the above sample run): QC v3 by Summary File: ALBH S Receiver type: rogue snr S 1 ################+++ A 2 #######++ S##S####### T 4 #############++ SI####++ E 5 ################++ L 6 #################+++ L 7 ###########++ S#### I 9 #############+++ S###I+++ T 12 *************++ SS++ S** E 14 ################++ 15 ###I+ S##S##########+ S## 16 #############++ S######IP++ 17 ################## ################ ##################P+ 20 ##############++ SI ##################++ 22 #I########++ S##########++ 23 ################# ###############++ 25 #####+++ S#############++ 26 #####++ S########+++ S###### 27 ##+++ S############## 28 ############+++ S###I###++ 29 ################## ###########++ S#########++ CLK :00 23:59 Time of First Epoch in File (year,month,day,hour) : : 0 Time of Last Epoch in File (year,month,day,hour) : :59 Observation Interval for File (in seconds) : 30 Elevation cutoff for qc : 15 Total number of observations expected : Total number of observations in file : Total number of points deleted : 2291 Data collection percentage : 100 RINEX vs qc point pos diff [Km] : 0.03 Average MP1 : Average MP2 : # of points for MP moving average : 50 Average clock drift [msec/hr] : Average time between resets [min] : Number of detected slips : 38 Observations per slip : 468 Page 18 Updated 3/7/95

14 2 : : : : : : : : : : : : : : : : : : : AVERAGE P2 MULTIPATH [M] : NUMBER OF OBSERV. USED FOR AVE : # POINTS FOR MP MOVING AVERAGE : 50 AVERAGE ELE 2 ANGLE [deg] : 43.0 NUMBER OF SLIPS <= 25 deg : 41 NUMBER OF SLIPS > 25 deg : 7 NUMBER OF RNX2 - SLIPS <= 25 deg : 32 NUMBER OF RNX2 - SLIPS > 25 deg : 5 NUMBER OF N - MSEC CLOCK SLIPS : 0 NUMBER OF DETECTED SLIPS : 38 NUMBER OF OBS / DETECTED SLIPS : NUMBER OF CODE L1 OBSERV. : NUMBER OF CODE L2 OBSERV. : RATIO OF P2 / P1 OBSERV. : CLOCK DRIFT [MSEC / HOUR] : 0.0 RECEIVER UNIT # : 292 ============================================= NORMAL COMPLETION Page 17 Updated 3/7/95

15 SLIP SLIP RXFL RXFL SAT-# #_of_obs. #_del AV_ELE P1_Multipath_[m] <_25 >_25 <_25 >_25 15 : : : : : : : : : : : : : : : : : : : : : : : : : AVERAGE P1 MULTIPATH [M] : NUMBER OF OBSERV. USED FOR AVE : # POINTS FOR MP MOVING AVERAGE : 50 AVERAGE ELE 1 ANGLE [deg] : 43.0 NUMBER OF SLIPS <= 25 deg : 41 NUMBER OF SLIPS > 25 deg : 7 NUMBER OF RNX1 - SLIPS <= 25 deg : 32 NUMBER OF RNX1 - SLIPS > 25 deg : 5 NUMBER OF N - MSEC CLOCK SLIPS : 0 NUMBER OF DETECTED SLIPS : 38 NUMBER OF OBS / DETECTED SLIPS : NUMBER OF CODE L1 OBSERV. : NUMBER OF CODE L2 OBSERV. : RATIO OF P2 / P1 OBSERV. : CLOCK DRIFT [MSEC / HOUR] : 0.0 RECEIVER UNIT # : 292 ============================================= SLIP SLIP RXFL RXFL SAT-# #_of_obs. #_del AV_ELE P2_Multipath_[m] <_25 >_25 <_25 >_25 15 : : : : : : Page 16 Updated 3/7/95

16 {A/S information. The satellites for which A/S was on for all or part of the time are displayed here.} ************************************************* A-S INFORMATION FOR RINEX FILE : ALBH O ************************************************* A-S WAS ON DURING THIS ARC AFFECTED SATELLITES AND TIMES ARE: SAT-# start Epoch start time stop epoch stop time 1 : : : : : : : : : : : : : : : : : : : : : : : : ************************************************* MULTIPATH + SLIP INFORMATION FOR RINEX FILE : ALBH O ************************************************* {A table showing detected slips and other information for each satellite in the RINEX observation file. The SAT-# column is the satellite id, the #_OF_OBS column is the number of complete observations for the SV, the #_del column is the number of points deleted for the SV, the AV_ELE column is the average elevation for all data points collected for the satellite, P1_Multipath_[m] is the average L1 P-Code multipath for the satellite, SLIP<_25 is the number of slips detected by QC at an elevation less than 25 (this elevation is specified in the qc.inp file), SLIP>_25 is the number of slips detected by QC that occurred above elevation 25 degrees, RXFL<_25 is the number of slips according to the RINEX file that were below 25 degrees and RXFL>_25 is the number of slips that occurred above 25 degrees according to the rinex file.} Page 15 Updated 3/7/95

17 {A table showing the number of points that were eliminated from calculations and output files. In order for QC to consider a data point, all available observables must be present.} # POINTS DELETED: ELEV CUT OFF: 2291 # POINTS DELETED: NO L1 PHASE : 0 # POINTS DELETED: NO L2 PHASE : 0 # POINTS DELETED: NO P1 CODE : 0 # POINTS DELETED: NO P2 CODE : 0 # POINTS DELETED: L1 low S/N : 0 # POINTS DELETED: L2 low S/N : 0 # POINTS DELETED: TOTAL : 2291 # POINTS TOTAL : # POINTS THAT WERE COMPLETE : NUMBER OF EPOCHS : 2880 {Clock information. Some receivers do not steer their oscillators to keep the receiver time near GPS time, but instead let the receiver time drift away from GPS time (Ashtech and Trimble receivers for example). However, in order to keep sampling data with in 1 ms of the true GPS time, the receiver will change it s sampling time by 1 ms. For example, if a GPS receiver of this type were collecting data at a rate of 30 seconds, the receiver would initially collect data at seconds and seconds. If the time in the receiver becomes more than 1 ms off from GPS time -- say the GPS receiver is 1ms fast, then the receiver will start collecting data at seconds and seconds. QC will look for these 1ms changes in the time tags and record them as clock drift events of 1 ms. Other receivers, such as the Turbo-Rouges, steer their internal oscillators so that the time inside the receiver is always well within 1 ms of GPS time. With these type of receivers, the clock drift terms will have no meaning since there will be no 1 ms changes in time GPS data is collected.} ************************************************* CLOCK + OSCILLATOR SUMMARY FOR RINEX FILE : ALBH O ************************************************* AVERAGE CLOCK DRIFT [MSEC/HOUR] : AVERAGE TIME BETWEEN RESETS [MIN] : Page 14 Updated 3/7/95

18 {Information about the time span, data collection interval and the data for each satellite is displayed. In the table below, the SV column is the ID if the satellite, L1 is the number of L1 phase values, L2 is the number of L2 phase values, P1 is the number of L1 P-Code values, P2 is the number of L2 P-Code values, CA is the number of L1 C/A values and #EXP is the number of data points expected which is calculated from the observation interval, elevation cut-off specified in the qc.inp file and the rise/set times of the SVs. If no orbit data is available, then the #EXP value will be zero.} RINEX FILE STARTS AT : IT ENDS AT : OBSERVATION INTERVAL [SEC] : SV L1 L2 P1 P2 CA #EXP FILE HAS P1 -PCODE DATA FILE HAS P2 -PCODE DATA FILE HAS C/A -PCODE DATA IN ADDITION TO P1 Page 13 Updated 3/7/95

19 CONVERTING ORBIT FILE FORMAT... FILE : ALBH N 225 ORBIT SETS TRANSLATED {The RINEX header of the observation file did not contain information about the number of SV s contained in the file, so the file is scanned for all SV s for which there is data. If the RINEX header contains this information, then this step is skipped.} NON-STANDARD RINEX HEADER - MUST SCAN ENTIRE FILE... 2 OBSERVATION DATA G (GPS) RINEX VERSION / TYPE RGRINEXN V2.1.0 VM EMR 20-FEB-95 05:00 PGM / RUN BY / DATE BIT 2 OF LLI (+4) FLAGS DATA COLLECTED UNDER AS CONDITION COMMENT HARDWARE CALIBRATION (S) COMMENT STATION INFORMATION UPDATED COMMENT ANT HGT.110 M BELOW L1 AND.128 M BELOW L2 PHASE CENTRE COMMENT albh WCDA-ACP VICTORIA, B.C., CANADA MARKER NAME AUTO-DOWNLOAD NATURAL RESOURCES CA OBSERVER / AGENCY 292 ROGUE SNR-8000 TURBO REC # / TYPE / VERS 368 DORNE MARGOLIN T ANT # / TYPE APPROX POSITION XYZ ANTENNA: DELTA H/E/N 1 1 WAVELENGTH FACT L1/2 5 P1 L1 L2 P2 C1 # / TYPES OF OBSERV 30 INTERVAL TIME OF FIRST OBS END OF HEADER Page 12 Updated 3/7/95

20 QC Output File: The QC program generates an extensive output file that is sent to the screen. This output file can be saved by redirecting it to a file, i.e. with the command: UNIX: % echo ALBH O > inp % echo AUTO N >> inp % qc < inp > ALBH0050.out Example output with annotations is listed below {comments are enclosed in brackets}: QC v3 05-JAN-1995 enter rinex file name (return for batch) --> enter rinex navigation file name (return for auto) --> {answer to above questions were supplied by the inp file} ===================================== QC vs 3; Last Modified : 05-JAN-1995 ===================================== ===================================== QC OF RINEX FILE : ALBH O USING ORBIT FILE : ALBH N OUTPUT PATH : ===================================== {QC echoes the parameters that were read in from the qc.inp file} PROCESSING PARAMETERS ARE : MAXIMUM IONOSPHERIC RATE (L1) CM/MIN :400 REPORT DATA GAP GREATER THAN MIN : 10 EXPECTED RMS LEVEL OF P1 MULTIPATH CM : 50 MULTIPATH SLIP SIGMA THRESHOLD SIGMA : 4 % INCREASE IN MP RMS FOR C/A & A/S :100 MOVING AVERAGE LEN FOR MP BIAS COR POINTS : 50 MINIMUM SIGNAL TO NOISE L1 : 0 MINIMUM SIGNAL TO NOISE L2 : 0 CONSIDER ORBIT FILE (Y=1,N=0) : 1 OUTPUT PLOT DECIMATED BY FACTOR OF : 1 DETECT CYCLE SLIPS (Y=1,N=0) : 1 ELEVATION CUTOFF FOR QC DEG : 15 COMPARISON ELEVATION DEG : 25 PREPARE PLOT FILE FOR MULTIPATH (Y=1,N=0) : 1 PREPARE PLOT FILE FOR L1 IONOSPHERE (Y=1,N=0) : 1 PREPARE PLOT FILE FOR L1 IONOSPHERE DOT (Y=1,N=0) : 1 PREPARE PLOT OF SAT AZIMUTH ANGLES (Y=1,N=0) : 1 PREPARE PLOT OF SAT ELEVATION ANGLES (Y=1,N=0) : 1 PREPARE SLIP OUTPUT FILE (Y=1,N=0) : 1 PLOT FLAG FOR CLOCK SWITCH EPOCHS (Y=1,N=0) : 1 ECHO RINEX OBS HDR INTO SUM FILE (Y=1,N=0) : 1 CONSIDER TIME WINDOW FILE (Y=1,N=0) : 0 ORBIT DATA SAMPLE PERIOD MIN : 10 DISPLAY WORKING PERCENTAGE (N=-1,Y=0 OR 6 (WRITE UNIT)) : 0 Page 11 Updated 3/7/95

21 Time Window File (qc.tim): If QC is to consider the time window file, then the file qc.tim (this file must be in the directory where QC was started) is opened and read. This file contains a start and end time between which data will be analyzed by QC. Any data that lies outside of this time window will then be ignored by QC. Whether or not QC uses this time window file is controlled by the settings in the qc.inp file. The format of the qc.tim file is: TIME WINDOW FILE FOR QC-PROGRAM =============================== YYYY MM DD HH MM SS.SSS #### <-- FROM <-- TO Page 10 Updated 3/7/95

22 Batch Mode File (qc.fil): Below is an example of a qc.fil file. The * s indicate the size of the input filed for the data line below. The first bit of information contained in this file is the path for output files. If you wish to have output go to the directory where QC was started, then this field can be left blank. At the end of the file, QC will expect to find the names for the RINEX observation file and RINEX ephemeris file, preceded by the set of qc.inp parameters to use. If the ephemeris file field is left blank, then QC will generate a RINEX ephemeris file name by changing the last character in the RINEX observation file to either n or N. Example qc.fil file: DATA FILES FOR GPS DATA QC PROGRAM SET PATH TO OUPUT DIRECTORY (no path: output=data directory; example: b:\out\) ******************************** path: SET RINEX DATA FILE RINEX EPHEMERIS FILE *** **************************************** **************************************** 1 kour o brdc n 1 albh o brdc n 2 zimm brdc n Page 9 Updated 3/7/95

23 CONSIDER TIME WINDOW FILE (Y=1,N=0) : 0 If this parameter is set to 1, then QC will open a file named qc.tim and read in a start and end time in which QC will evaluate data. The format of this file is explained in the section Time Window File. ORBIT DATA SAMPLE PERIOD MIN : 10 In order to speed up the time it takes to run the QC program, satellite positions are only calculated at the interval specified here. To obtain a SV position at any given time, the positions taken at times that bracket the desired time are interpolated. DISPLAY WORKING PERCENTAGE (N=-1,Y=0 OR 6 (WRITE UNIT)) : -1 If set to either 0 or 6, then QC will display a status character and a display of how much data has been analyzed. The number specified is the FORTRAN UNIT number the data will be written to. For UNIX systems, UNIT 0 is connected to stderr, which means that if QC output is redirected into a file, the status information will still appear on the screen instead of in the redirected file. For DOS and VMS, this is not the case and there is no real difference between setting the value to 0 or 6. If set to -1, then QC will not display any status information. Page 8 Updated 3/7/95

24 OUTPUT PLOT DECIMATED BY FACTOR OF : 1 This value determines if the output plot files should be decimated or not. A value of 1 will cause all data points to be written to output plot files, 2 will cause every other point to be written, 3 every third point and so on. DETECT CYCLE SLIPS (Y=1,N=0) : 1 If set to 1, then QC will look for cycle slips on the L1 and L2 phases. ELEVATION CUTOFF FOR QC DEG : 15 QC will not consider any data from a satellite whenever it is below this elevation cut off angle. Note that if no orbit information is available, or is not requested, then QC will set the elevation angle for all SV s to 90 degrees and this parameter will have no effect. COMPARISON ELEVATION DEG : 25 QC will tabulate data such as the number of cycle slips in two separate groups for each satellite -- one for data points collected while the satellite was below this elevation and another for data points collected while the satellite was above this elevation. PREPARE PLOT FILE FOR MULTIPATH (Y=1,N=0) : 1 If set to 1, then plot output files for L1 code multipath (.mp1 file) and L2 code multipath (.mp2 file) will be generated. If set to 0, these files will not be generated. PREPARE PLOT FILE FOR L1 IONOSPHERE (Y=1,N=0) : 1 If set to 1, then a plot file containing the ionosphere values (.ion) will be generated. PREPARE PLOT FILE FOR L1 IONOSPHERE DOT (Y=1,N=0) : 1 If set to 1, then a plot file containing the derivative of the ionosphere values (.iod) will be generated. PREPARE PLOT OF SAT AZIMUTH ANGLES (Y=1,N=0) : 1 If set to 1 then a plot file with satellite azimuth angles (.azi) will be generated PREPARE PLOT OF SAT ELEVATION ANGLES (Y=1,N=0) : 1 If set to 1 then a plot file with satellite elevation angles (.ele) will be generated. PREPARE SLIP OUTPUT FILE (Y=1,N=0) : 1 If set to 1 then an output file listing slips (.slp) will be generated. ECHO RINEX OBS HDR INTO SUM FILE (Y=1,N=0) : 1 If set to 1 then the header of the RINEX observation file will be echoed in to the summary output file. PLOT FLAG FOR CLOCK SWITCH EPOCHS (Y=1,N=0) : 1 Create flags in output plot files if there is a clock slip (C flag). Page 7 Updated 3/7/95

25 QC Input Parameters: MAXIMUM IONOSPHERIC RATE (L1) CM/MIN :400 The maximum the ionosphere is allowed to change in units of cm/min. QC will calculate the ionospheric delay of the GPS L1 and L2 phase measurements by forming a linear combination. If this value changes from one epoch to the next by more that the rate specified, then the epoch will be marked as a cycle slip. REPORT DATA GAP GREATER THAN MIN : 10 If there is a data gap in the input RINEX file (no RINEX data records) greater than this value, then QC will display a row of G s in the summary output file (.yrs summary file). EXPECTED RMS LEVEL OF P1 MULTIPATH CM : 50 This is the expected RMS value of P-Code multipath. This is used in conjunction with the multipath slip sigma threshold (next parameter) to determine multipath slips. As QC calculates multipath values, it keeps track of the running average and RMS for each SV. If the RMS value is less than the value here, then this value is multiplied by the sigma to obtain the amount the multipath must change by to be considered a slip. If the current running RMS is greater, then it will be used instead of this value. A setting of 999 will cause QC to not look for jumps in the MP values. MULTIPATH SLIP SIGMA THRESHOLD SIGMA : 4 This value is used with the above parameter to determine how much multipath must change in order to be considered a slip. Above is a more detailed explanation. % INCREASE IN MP RMS FOR C/A & A/S :100 If multipath calculations are made with C/A or while A/S is on, then the expected RMS level of P1 and P2 (if present) multipath is increased by this percentage. MOVING AVERAGE LEN FOR MP BIAS COR POINTS : 50 If this value is non-zero, than then a moving average of multipath values will be calculated and removed from each point calculated for each satellite. If there is a data gap (see second parameter) then the moving average will be reset to zero. This value acts as a simple high pass filter and effects both the average RMS values reported by QC and the multipath plot files. A value of 0 will disable this option and 100 is the maximum value. MINIMUM SIGNAL TO NOISE L1 : 0 MINIMUM SIGNAL TO NOISE L2 : 0 These values determine the minimum signal to noise ratio that L1 and L2 signals must have in order be considered by QC as a valid data point. A value of 0 will accept all data. CONSIDER ORBIT FILE (Y=1,N=0) : 1 If this value is set to 1, then QC will look for an orbit file to use for calculating satellite positions (azimuth and elevation angles). If QC is not explicitly given a RINEX navigation file name, then QC will look for a file with the same name as the RINEX observation file but ending with either a n or N. If this value is set to 0, then QC will not calculate azimuth or elevation data for the satellite positions. It will use an azimuth value of 0 degrees and an elevation value of 90 degrees for all SV s. Page 6 Updated 3/7/95

26 COMPARISON ELEVATION DEG : 25 PREPARE PLOT FILE FOR MULTIPATH (Y=1,N=0) : 0 PREPARE PLOT FILE FOR L1 IONOSPHERE (Y=1,N=0) : 0 PREPARE PLOT FILE FOR L1 IONOSPHERE DOT (Y=1,N=0) : 0 PREPARE PLOT OF SAT AZIMUTH ANGLES (Y=1,N=0) : 0 PREPARE PLOT OF SAT ELEVATION ANGLES (Y=1,N=0) : 0 PREPARE SLIP OUTPUT FILE (Y=1,N=0) : 0 PLOT FLAG FOR CLOCK SWITCH EPOCHS (Y=1,N=0) : 1 ECHO RINEX OBS HDR INTO SUM FILE (Y=1,N=0) : 1 CONSIDER TIME WINDOW FILE (Y=1,N=0) : 0 ORBIT DATA SAMPLE PERIOD MIN : 5 DISPLAY WORKING PERCENTAGE (N=-1,Y=0 OR 6 (WRITE UNIT)) : 0 Notice that there can be more than one set of parameters. If QC is run in its interactive mode by typing the name of the RINEX file to analyze directly into QC, then the first (default) set of parameters are used. If, however, QC is run in its batch mode, then QC will use the data set indicated in the qc.fil input file. For more information, see the explanation of the qc.fil file. Page 5 Updated 3/7/95

27 Configuration File: The file qc.inp is required to be in the directory where QC is started and contains parameters that QC will use to determine the data quality and which output files should be generated. A sample qc.inp file is shown below: INPUT PARMATERS FOR GPS QC PROGRAM: #### SET#: 1 (DEFAULT SET) **************************************** *** MAXIMUM IONOSPHERIC RATE (L1) CM/MIN :400 REPORT DATA GAP GREATER THAN MIN : 10 EXPECTED RMS LEVEL OF P1 MULTIPATH CM : 50 MULTIPATH SLIP SIGMA THRESHOLD SIGMA : 4 % INCREASE IN MP RMS FOR C/A & A/S :100 MOVING AVERAGE LEN FOR MP BIAS COR POINTS : 50 MINIMUM SIGNAL TO NOISE L1 : 0 MINIMUM SIGNAL TO NOISE L2 : 0 CONSIDER ORBIT FILE (Y=1,N=0) : 1 OUTPUT PLOT DECIMATED BY FACTOR OF : 1 DETECT CYCLE SLIPS (Y=1,N=0) : 1 ELEVATION CUTOFF FOR QC DEG : 10 COMPARISON ELEVATION DEG : 25 PREPARE PLOT FILE FOR MULTIPATH (Y=1,N=0) : 1 PREPARE PLOT FILE FOR L1 IONOSPHERE (Y=1,N=0) : 1 PREPARE PLOT FILE FOR L1 IONOSPHERE DOT (Y=1,N=0) : 1 PREPARE PLOT OF SAT AZIMUTH ANGLES (Y=1,N=0) : 1 PREPARE PLOT OF SAT ELEVATION ANGLES (Y=1,N=0) : 1 PREPARE SLIP OUTPUT FILE (Y=1,N=0) : 1 PLOT FLAG FOR CLOCK SWITCH EPOCHS (Y=1,N=0) : 1 ECHO RINEX OBS HDR INTO SUM FILE (Y=1,N=0) : 1 CONSIDER TIME WINDOW FILE (Y=1,N=0) : 0 ORBIT DATA SAMPLE PERIOD MIN : 10 DISPLAY WORKING PERCENTAGE (N=-1,Y=0 OR 6 (WRITE UNIT)) : 0 #### SET#: 2 **************************************** *** MAXIMUM IONOSPHERIC RATE (L1) CM/MIN :900 REPORT DATA GAP GREATER THAN MIN : 30 EXPECTED RMS LEVEL OF P1 MULTIPATH CM :100 MULTIPATH SLIP SIGMA THRESHOLD SIGMA : 5 % INCREASE IN MP RMS FOR C/A & A/S :200 MOVING AVERAGE LEN FOR MP BIAS COR POINTS : 50 MINIMUM SIGNAL TO NOISE L1 : 0 MINIMUM SIGNAL TO NOISE L2 : 0 CONSIDER ORBIT FILE (Y=1,N=0) : 1 OUTPUT PLOT DECIMATED BY FACTOR OF : 1 DETECT CYCLE SLIPS (Y=1,N=0) : 1 ELEVATION CUTOFF FOR QC DEG : 15 Page 4 Updated 3/7/95

28 covered in detail later. In this example we typed in the name of the RINEX observation file that QC is to process. Next, QC asks for the name of the RINEX navigation file to read. If we had not entered a file name for the RINEX observation file, then QC would have read the qc.fil file to get the name of the navigation file instead of prompting the user for the file name. If no file name is entered for the rinex navigation file, then QC will look for a file with the same name as the RINEX observation file with an N (QC tries both an upper and lower case N) as the last character. If neither of these files exists, as is the case here, then QC will not try to read orbit information. After file names have been specified, QC starts to process the file and sends detailed output to the standard output device. A convenient way to run QC so that the exhaustive output will be saved in a file is to place the name of the RINEX file to process in a file, say inp.txt, and run QC with standard input redirected from a file and standard output redirected to a file. For example: UNIX: tonga:/u3/jjohnson/qc% echo math o > inp.txt tonga:/u3/jjohnson/qc% echo brdc n >> inp.txt tonga:/u3/jjohnson/qc% qc < inp.txt > math0310.out DOS: C:\PGM\QC> ECHO MATH O > INP.TXT C:\PGM\QC> ECHO BRDC N >> INP.TXT C:\PGM\QC> QC < INP.TXT > MATH0310.OUT VMS/DCL: $ OPEN/WRITE TMPFIL INP.TXT $ WRITE TMPFIL MATH O $ WRITE TMPFIL BRDC N $ CLOSE TMPFIL $ DEFINE/USER SYS$INPUT INP.TXT $ DEFINE/USER SYS$OUTPUT MATH0310.OUT $ RUN QC To run QC in a batch mode using the qc.fil file to specify file names, a single blank line is placed in the inp.txt file. One advantage of using the qc.fil file for input is that more than one file at a time can be specified. Page 3 Updated 3/7/95

29 Getting Started: QC is available on several computing platforms and the executable for QC has a different name for each one. Below is a list of the currently available versions of QC: Table 1: Different QC Versions Available Executable Name Platform QC_MSF51.EXE MS-DOS compiled with Microsoft Fortran 5.1. QC_LAHEY.EXE MS-DOS with DOS extender compiled with Lahey Fortran. QC_SUN4_413 SUN-4 under SunOS QC_SUN4_53 SUN-4 under Solaris 5.3 QC_VAX.EXE QC_AXP.EXE DEC-VAX under VMS DEC-Alpha under VMS QC_HPUX_91 HP under UNIX 9.1 QC_AIX_3 IBM AIX Version 3 Once you have obtained the correct version for your system, you can rename the QC executable to QC (or QC.EXE if the.exe is needed) for conveinance. In order to use QC, you will need to have at least three files: the qc executable file, a qc.inp file (input parameter file), and a RINEX observation file. The simplest way to run QC is to simply run the program from the command prompt and then type in the name of the RINEX file to process. For example (QC_SUN4_53 was renamed to qc for all examples): tonga:/u3/jjohnson/qc% qc enter rinex file name (return for batch) --> BLMB O enter rinex navigation file name (return for auto) --> ===================================== QC vs 3; Last Modified : 31-OCT-1994 ===================================== ===================================== QC OF RINEX FILE : BLMB O USING ORBIT FILE : OUTPUT PATH : ===================================== The above output shows just the first part of the output from QC. The characters in bold are what the user has to type in. In this case, we only have a RINEX observation file named BLMB O and no RINEX navigation file. The first question QC will ask is the name of the RINEX observation file to process. If no file name is entered, then QC will assume it is to run in batch mode and read the file qc.fil for the names of the files to process. The qc.fil format is Page 2 Updated 3/7/95

30 QC v3 Users Guide Document Date: March 1994 University NAVSTAR Consortium P. O. Box 3000 Boulder, CO USA (TEL) (FAX) Introduction: QC is a program for quality checking static and kinematic GPS data. This program checks the data from a single station and is available to run on 286 or higher PC computers and on a variety of other platforms such as UNIX workstations. The program forms linear combinations of the GPS range and carrier phase data to compute (1) L1 pseudorange multipath for C/A- or P-code observations, (2) L2 pseudorange multipath for P-code observations, (3) Ionospheric phase effects on the L1 carrier frequency, and (4) The rate of change of the ionospheric delay. The program also writes data summary files with information about the cycle slips, receiver multipath, satellite elevation angles, receiver clock drift, etc. Data from any GPS receiver can be quality checked if they are in RINEX format. If satellite position information is to be calculated and used by QC, then a RINEX navigation file must also be available. RINEX translators developed at the University of Bern are publicly available and distributed with the QC programs. There are also batch and script files that translate raw data to RINEX format, processes the data with QC, then displays the output plot files using a graphical display tool (either QCVIEW or GT) from a single command line. Batched runs can generate hard copies of the QC summary and graphics files. This is useful to check a large number of data in batch mode. The purpose of this documentation is to explain the details of the QC program and its output. The details of RINEX conversion programs, batch and script files and viewing programs (which are documented elsewhere) will not be discussed. Acknowledgments: QC was originally written by Dr. Chris Rocken at and has been extensively modified by Teresa Van Hove, John Braun and James Johnson. The staff of, among others, played a key role in trouble shooting problems and suggesting improvements. QC contains some code written at the University of Bern (RINEX readers, Orbits, etc.) by their permission. Page 1 Updated 3/7/95