pdf.html

 TOC  |      | Pages | Gpscom | Next Section | INDEX

Problem Definition File

The problem definition file is a required control file and, it is currently required to be named gpscom.pdf and to be in the directory from which the program gpscom is run. This file defines the various options that are to be followed by the program as well as the constraints that are to be applied to the global adjustment. gpscom.pdf is an ASCII file and, therefore, easily modifiable with an editor.

The very first line of this file is required to specify the name of a site information file. This appears first partly for historical reasons and partly because of the importance of the site information file and the requirement that at least one is available. The site information file is designed to provide a complete description and history of a tracking site including all apriori information. Additional site information files can be supplied if needed, see STATION INFO below.

The options section follows the site information file. This section is very close to the same as the Constraint Control Section of the pages.skl file used by the Pages program. Many of the constraints and options are the same and specified in the same way however, there are also several options that are unique to each program. All of the options available to the Gpscom program are listed below.

The problem definition file requires two identifiers regardless of the inclusion of any other identifiers:

OPTIONS: This identifies the start of the problem definition file options section.
END_OPTS: This identifies the end of the problem definition file options.

In general all other constraint identifiers are optional however, some subset of the possible options will be necessary to get a satisfactory solution. For example some choice of atleast minimal constraints is needed. Because the problem definition file identifiers can appear in any order, the various possible identifiers will be listed alphabetically with their descriptions and examples, where necessary. The case, spelling and spacing of these identifiers is significant. Some require additional information. These cases will be noted and described in detail including FORTRAN format specifications.

Following is a list, in alphabetical order, of all current options available in the problem definition file of the Gpscom program:

ANTENNA PHASE CORR

Use the antenna phase correction file specified on the following line to determine apriori offset information for the tracking sites(see, brief description or antenna calibration project).
NOTE: requires additional input on next line.
The identifier is followed by a file name.
FORMAT: (A)

AVG. SITE CONST
End is identified by END-AVSI

Constrains the centroid of a group of sites. Each site could shift but the average of the coordinates must not.
NOTE: requires additional input on subsequent lines.
There are two possible options to apply with this constraint:

1. USE A-PRIORI VALUES
This option causes Gpscom to use the centroid of the input station coordinates in defining the constraint equations. The next line must give a variance to be applied to the constraint equation.
UNITS: meters squared
FORMAT: (13X, F16.6)
Immediately following the variance line are lines designating the sites to be constrained, one line per site.

Example:

           AVG. SITE CONST    
           USE A-PRIORI VALUES
                                  1.0e-06
           sant a  5
           mcmu a  4
           END-AVSI

2. If USE A-PRIORI VALUES is not specified, then the variance line is followed by the lines designating the sites to be constrained and the coordinates to use for defining the centroid of those sites, one site per line.
UNITS: meters
FORMAT: (A12, 1X, 3F16.6)

Example:

           AVG. SITE CONST    
                                  1.0e-06
           sant a  5      1769693.285393  -5044574.219096  -3468321.218509
           mcmu a  4     -1310696.280131    310469.128906  -6213368.431436
           END-AVSI

AVG. SITE VEL. CONST.
End is identified by END-SITE

Constrains the velocity centroid of a group of sites. Each site's velocity could shift, but the average must not.
NOTE: requires additional input on subsequent lines.
The next line must define a constraint on either rotation or translation, and both may be included, one immediately following the other. These options are defined as follows:
ROTATION
Indicates a constraint on the rotation of the network, as defined by the velocities of the constrained sites, plus three rotational velocity components to be applied and a variance for the constraint.
UNITS: radians/year for velocity; (radians/year)squared for variance
FORMAT: (A12, 1X, 4F16.8)
TRANSLATION
Indicates a constraint on the translation of the network, as defined by the velocities of the constrained sites, plus three velocity components to be applied and a variance for the constraint.
UNITS: meters/year for velocity; (meters/year)squared for variance
FORMAT: (A12, 1X, 4F16.8)
Following these entries, the identifiers of the sites to be used in the constraint are given, one entry per line.
Note that the velocity components defined in these constraints may be zero, or can be computed from an a-priori velocity model for the sites selected. To do so one must get the sum of the cross products of the (A-priori velocity - Model velocity) times the coordinate vector for the sites to be included.

Example:

    AVG. SITE VEL. CONST.
    TRANSLATION             0.000      0.000     0.000     1.000e-05
    ROTATION                0.000      0.000     0.000     1.000e-05
    sant a  5
    mcmu a  4
    END-SITE

COMMENTS
End is identified by END_COMM

Any remarks desired may be placed between these identifiers.

CONSTRAIN OPR RAD PR1: Constrain the scale factors for the once-per-revolution radiation pressure 1 terms. The sigma, in dimensionless units, is also entered in this block.
NOTE: requires additional input on next line.
The identifier is followed by the standard deviation to be applied.
UNITS: seconds
FORMAT: free format

CONSTRAIN OPR RAD PR2: Constrain the scale factors for the once-per-revolution radiation pressure 2 terms. The sigma, in dimensionless units, is also entered in this block.
NOTE: requires additional input on next line.
The identifier is followed by the standard deviation to be applied.
UNITS: seconds
FORMAT: free format

CONSTRAIN OPR RAD PR3: Constrain the scale factors for the once-per-revolution radiation pressure 3 terms. The sigma, in dimensionless units, is also entered in this block.
NOTE: requires additional input on next line.
The identifier is followed by the standard deviation to be applied.
UNITS: seconds
FORMAT: free format

CONSTRAIN RAD PR1

Constrain the radiation pressure 1 scale factors. The sigma, in dimensionless units, is also entered in this block.
NOTE: requires additional input on next line.
The identifier is followed by the standard deviation to be applied.
UNITS: dimensionless
FORMAT: free format

CONSTRAIN RAD PR2: Constrain the radiation pressure 2 scale factors. The sigma, in dimensionless units, is also entered in this block.
NOTE: requires additional input on next line.
The identifier is followed by the standard deviation to be applied.
UNITS: dimensionless
FORMAT: free format

CONSTRAIN RAD PR3: Constrain the radiation pressure 3 scale factors. The sigma, in dimensionless units, is also entered in this block.
NOTE: requires additional input on next line.
The identifier is followed by the standard deviation to be applied.
UNITS: dimensionless
FORMAT: free format

CONSTRAIN SAT POS: Constrain the adjustments to piece-wise linear satellite coordinates to require, within limits of the sigma, that the successive values are equal. With this method using different values for sigma allows one to dampen the rate of change of the adjusted parameters, allowing flexibility, but not too much. The sigma, in meters, is also entered in this block.
NOTE: requires additional input on next line.
The identifier is followed by the standard deviation to be applied.
UNITS: meters
FORMAT: free format

CONSTRAIN SAT VEL: Constrain the adjustments to piece-wise linear satellite velocities to require, within limits of the sigma, that the successive values are equal. With this method using different values for sigma allows one to dampen the rate of change of the adjusted parameters, allowing flexibility, but not too much. The sigma, in meters, is also entered in this
The identifier is followed by the standard deviation to be applied.
UNITS: meters
FORMAT: free format

CONSTRAIN AVERAGE UT1

Constrains the average UT1 estimates.
NOTE: requires additional input on next line.
The identifier is followed by the standard deviation to be applied.
UNITS: seconds
FORMAT: free format

CONSTRAIN FIRST UT1

Constrains the first UT1 estimate. When this option is used and several data sets are being combined with the piece wise linear model, which is the default action, then exactly one UT1 constraint is applied. However in the presence of the option "DAILY EOP VALUES", see below, then the first UT1 parameter in each data set will be constrained.
NOTE: requires additional input on next line.
The identifier is followed by the standard deviation to be applied.
UNITS: seconds
FORMAT: free format

CONST. INDIVIDUAL SITE
End is identified by END-INDV-SITES

Constrains the coordinates of the indicated sites.

NOTE: requires additional input on subsequent lines.

There are two possible options to apply with this constraint:

1. USE A-PRIORI VALUES

This option causes Gpscom to use the a-priori station coordinates and sigmas to define the constraints. The site ID's and an additional factor which is multiplied times the weight to be applied to the constraints follow, one entry per line.

UNITS: meters squared

FORMAT: (A12, 1X, E16.8)

Example:

         CONST. INDIVIDUAL SITE
         USE A-PRIORI VALUES
         kokb a  9         1.000e+00
         algo a  7         1.000e+00
         madr a  7         1.000e+00
         END-INDV-SITES

2. Specify the site ID's, coordinates, and variances to be applied.

UNITS: meters for coordinates; meters squared for variance

FORMAT: (A12, 1X, 3F16.8, E16.8)

Example:

         CONST. INDIVIDUAL SITE
         sant a  5     1769693.285393 -5044574.219096 -3468321.218509  1.000e-10
         mcmu a  4    -1310696.280131   310469.128906 -6213368.431436  1.000e-10
         END-INDV SITES

CONST. INDIVIDUAL VEL.
End is identified by END-INDV-VEL.

Constrains the indivdual sites' velocities.

NOTE: requires additional input on subsequent lines.

There are two possible options to apply with this constraint.

1. USE A-PRIORI VALUES

This option causes Gpscom to use the input station velocities and sigmas as constraints. The site ID's and an additional factor which is multiplied times the weight to be applied to the constraints follow, one entry per line.

UNITS: (meters/year)squared

FORMAT: (A12, 1X, E16.8)

Example:

         CONST. INDIVIDUAL VEL.
         USE A-PRIORI VALUES
         kokb a  9         1.000e+00
         algo a  7         1.000e+00
         madr a  7         1.000e+00
         END-INDV-VEL.

2. Specify the site ID's, velocities, and variances to be applied.

UNITS: meters/year for velocities; (meters/year)squared for variance.

FORMAT: (A12, 1X, 3F16.8, E16.8)

Example:

         CONST. INDIVIDUAL VEL 
         sant a  5           0.005          -0.007           0.002     1.000e-05
         mcmu a  4           0.110          -0.110          -0.103     1.000e-05
         END-INDV-VEL.

DAILY EOP VALUES

This option causes the program Gpscom to change the model for the EOP parameters to compute independent daily values of EOP at the begining and end of each daily data set rather than the default model of piece wise linear terms from day to day. This option will change the effect of the "CONSTRAIN FIRST UT1" option, see above, so that the first UT1 parameter in each data set will be constrained.

EQUAL VELOCITY PAIRS
End is identified by END-PAIR

This identifier forces pairs of sites to have the same velocity.
NOTE: requires additional input on subsequent lines.
The next line must give the variance to be applied as the constraint.
UNITS: (meters/year)squared
FORMAT: (13X, F16.8)
Subsequent lines contain two site ID's per line, designating the
site pairs to constrain.
FORMAT: (A12, 2X, A12)

Example:

         EQUAL VELOCITY PAIRS
                               1.0e-07
         gait a  3     gode a  4
         kosg a  7     brus a  9
         END-PAIR

GOOGE NUMBERS

Print Googe numbers in the global print file.

HOLD APRIORI SCALE

Constrains the apriori scale of the reference frame. The sites used to define this constraint are the same used in the list for either identifier AVG. SITE CONST. or CONST. INDIVIDUAL SITE. Constrains the adjustment such that the sites move inwards or outwards from the center of the frame so that the original scale is unchanged.
NOTE: requires additional input on next line.
The identifier is followed by the variance to be applied.
UNITS: (radians/year)squared
FORMAT: (13X, F16.8)

HOLD APRIORI VELOCITIES

Constrains all sites to their apriori velocities.
NOTE: requires additional input on next line.
The identifier is followed by the standard error to be applied.
UNITS: (radians/year)squared
FORMAT: (13X, F16.8)

KNOWN SITE PAIRS
End is identified by END-PAIR

This identifier causes the program to apply constraints to force pairs of sites to have the final coordinates which are offset by an amount specified in this section. The offset specified is the known x,y,x difference between the site positions in the sence of the first site coordinates minus the second sites coordinates.
NOTE: requires additional input on subsequent lines.
The next line must give the standard error to be applied as the constraint.
UNITS: meters
FORMAT: (13X, F16.8)
Subsequent lines contain two site ID's per line, designating the
site pairs to constrain, followed by the x,y,z offsets to be enforced between the sites coordinates.
FORMAT: (A12,2x,A12,3F10.5)

Example:

         KNOWN SITE PAIRS
                               1.0e-07
         gait a  3     gode a  4        .0500     .0450     .0100
         kosg a  7     brus a  9        .0000     .0000     .0000
         END-PAIR

LOCAL POLE-UT1

Causes the earth orientation parameters to be forward reduced and eliminated from the adjustment as nuisance parameters and thus not carried forward into the global part of the output normal equation file. Adjusted results will be printed from the current global solution and will be retained in the normal equation as local parameters.

LOCAL SATELITES

Causes the satelite parameters to be forward reduced and eliminated from the adjustment as nuisance parameters and thus not carried forward into the global part of the output normal equation file. Adjusted results will be printed from the current global solution and will be retained in the normal equation as local parameters.

LOCAL TROPOSPHERE

Causes the troposphere parameters to be forward reduced and eliminated from the adjustment as nuisance parameters and thus not carried forward into the global part of the output normal equation file. Adjusted results will be printed from the current global solution and will be retained in the normal equation as local parameters.

MID_POINT COORDINATES

Print site coordinates at the mid-point of the data span.

NO APRIORI INFO MATRIX

This option instructs gpscom not to include the aprioir information matrix in the Sinex file.

NO XYZ ROTATION ABOUT X

Constrains the average rotation of the network as defined by the computed coordinate shifts, about the X-axis. The sites used to define this constraint are the same used in the list for either identifier AVG. SITE CONST. or CONST. INDIVIDUAL SITE.
NOTE: requires additional input on next line.
The identifier is followed by the variance to be applied.
UNITS: (radians/year)squared
FORMAT: (13X, E16.8)

NO XYZ ROTATION ABOUT Y

Constrains the average rotation of the network as defined by the computed coordinate shifts, about the Y-axis. The sites used to define this constraint are the same used in the list for either identifier AVG. SITE CONST. or CONST. INDIVIDUAL SITE.
NOTE: requires additional input on next line.
The identifier is followed by the variance to be applied.
UNITS: (radians/year)squared
FORMAT: (13X, E16.8)

NO XYZ ROTATION ABOUT Z

Constrains the average rotation of the network as defined by the computed coordinate shifts, about the Z-axis. The sites used to define this constraint are the same used in the list for either identifier AVG. SITE CONST. or CONST. INDIVIDUAL SITE.
NOTE: requires additional input on next line.
The identifier is followed by the variance to be applied.
UNITS: (radians/year)squared
FORMAT: (13X, E16.8)

PROJECT NAME

The project name.
NOTE: requires additional input on next line.
FORMAT: (A72)

PROJECTED EPOCH

The coordinates of all sites are rotated to a common epoch on output. The default is the start of the year the data were taken. This default can be changed with this identifier.
NOTE: requires additional input on next line.
The identifier is followed on the next line by a date.
FORMAT: one of the following:

year month day hour minute second
year day-of-year hour minute second
year month day
year day-of-year
modified Julian date

STATION INFO

Additional station information file names.
NOTE: requires additional input on next line.
The identifier is followed by a file name.
FORMAT: (A)

TIME FOR APRIORI VALUES

Occasionally, the coordinates for a site change, when the reference frame is updated, for example. Gpscom searches the site information for the appropriate coordinates for the date the data were taken. This identifier limits the search effectively forcing sites to have consistent coordinates through all processing. This is essential for long term projects.
NOTE: requires additional input on next line.
The identifier is followed by a date.
FORMAT: one of the following:

year month day hour minute second
year day-of-year hour minute second
year month day
year day-of-year
modified Julian date

TOLERANCE

The tolerance to singularities. LS matrix rows with Googe numbers smaller than this value will be designated singular and set to zero. When the tolerance is not specified in this manner the value 1.0d-5 will be used as a default.
NOTE: requires a value on next line.
UNITS: dimensionless
FORMAT: (E10.3)

TURN OFF AVG. SITE X

Turns off the constraint on the average X coordinate for the sites selected in identifier AVG. SITE CONST.

TURN OFF AVG. SITE Y

Turns off the constraint on the average Y coordinate for the sites selected in identifier AVG. SITE CONST.

TURN OFF AVG. SITE Z

Turns off the constraint on the average Z coordinate for the sites selected in identifier AVG. SITE CONST.

USE PHASE PATTERN OFFSETS

The IGS format antenna phase correction file, ant_info, contains the appropriate offsets from the antenna reference point to the L1 and L2 phase centers. This identifier forces Gpscom to use this information. This is the default action of the program.
Alternatively, there is another constraint identifier which will override this default action. This identifier is given as: <\DL>

DO NOT USE PHASE PATTERN OFFSE This identifier forces Gpscom to ignore the information in the IGS format antenna phase correction file, ant_info, regarding the offsets from the antenna reference point to the L1 and L2 phase centers.

USE PHASE PATTERN HORIZONTAL O

The IGS format antenna phase correction file, ant_info, contains the appropriate offsets from the antenna reference point to the L1 and L2 phase centers. This identifier forces Gpscom to use the horizontal information. This is the default action of the program.
Alternatively, there is another constraint identifier which will override this default action. This identifier is given as:

DO NOT USE PHASE PATTERN HORIZ This identifier forces Gpscom to ignore the information on the horizontal components in the, ant_info, IGS format antenna phase correction file regarding the offsets from the antenna reference point to the L1 and L2 phase centers.

USE PHASE PATTERN VERTICAL OFF

The IGS format antenna phase correction file, ant_info, contains the appropriate offsets from the antenna reference point to the L1 and L2 phase centers. This identifier forces Gpscom to use the vertical information. This is the default action of the program.
Alternatively, there is another constraint identifier which will override this default action. This identifier is given as:

DO NOT USE PHASE PATTERN VERTI This identifier forces Gpscom to ignore the information on the vertica components in the IGS format antenna phase correction file, ant_info, regarding the offsets from the antenna reference point to the L1 and L2 phase centers.

WRITE NORMAL MATRIX

Causes Gpscom to create the normal matrix output file named pag4.nrm. Several of these files for different data sets can then be joined into a combined adjustment of the global parameters using the program gpscom.

WRITE SINEX FILE

Causes Gpscom to create the SINEX output file named gpscom.snx.
NOTE: requires additional input on subsequent lines.
This identifier requires the following information, one entry per line, in this exact order:

DESCRIPTION: description of the processing which produced this SINEX file;
OUTPUT: organization creating this file;
CONTACT: email address of person to contact with questions regarding this processing;
SOFTWARE: program name;
HARDWARE: system/hardware description;
INPUT: data source.

FORMAT: (A16, A)

Unique Keys

Some Keys Unique to Gpscom

 MID-POINT COORDINATES
 GOOGE NUMBERS
 ANTENNA PHASE CORR
 PROJECTED EPOCH

Some Keys Unique to Pages


 MAKE SITE TROPO GLOBAL      or    DEFINE JUNCTION SITES
 END_TROPO                   or    END-JUNCT
 MAKE SATELITES GLOBAL
 MAKE EOP GLOBAL
 NO SOLUTION
 APPLY DD CORRELATIONS
 OBSERVATIONS STANDARD ERROR
 TROPO CONSTR
 OUTPUT INTEGER FILES
 AGENCY ID
 FRAME ID
 EPOCH FOR OUTPUT

Sample gpscom.pdf File

 TOC  |      | Pages | Gpscom | Next Section | INDEX

pdf.html
March 4, 1999
Bill Dillinger