The package encompasses various GNSS-related functionality such as efficient reading and writing GNSS files (e.g. SINEX, SP3, CLK, IONEX and many others), advanced analysis and comparison, various coordinate transformations including geodetic frame rotations, predictions and combinations. Package Solver.
pip install gnssanalysis
- BLQ
- BSX/BIA
- CLK
- ERP
- IONEX
- NANU
- RINEX
- SINEX (including discontinuity, post-seismic file formats)
- SP3
- TROP
- GINAN proprietary formats: PEA partials, POD output, STEC and TRACE
There is a set of standalone utilities installed together with the module module which are build on top of gnssanalysis.
A utility originally created for automated testing of the computed GNSS files relative to the known good solution files.
The simplest use case for the diffutil
is to call it with two GNSS files specified after -i
:
diffutil -i file1 file2
diffutil
parses files' extensions and automatically calls a command needed, e.g. sp3
for file.sp3
. It is also possible to specify the command manually in case file extensions are non-standard or missing.
diffutil -i file1 file2 sp3 # sp3 is a command inside diffutil
Reads any number of sinex files given, specifically the SITE/ID block and creates an interactive html map with all the stations plotted. Every file will get a unique color marker, with decreasing size for each additional marker, constructing "lollipops" at common stations. This allows seeing intersections of stations within files to be easily seen. The sinex files may also be compressed (either .Z or .gz)
How to use:
snxmap snxfile1 snxfile2.Z snxfile3.gz
-o path_to_output
Merges any number of sp3 files together creating sp3 file of any length. Could also accept clk files to populate merged sp3 with clock offset values.
How to use:
sp3merge -s file1.sp3 file2.sp3.Z file3.sp3.gz -c file1.clk file2.clk.Z file3.clk.gz
A utility to parse collection of igs-format station files and create a sinex file with required station information - location, station hardware etc.
How to use:
log2snx -l "~/logfiles/*/*log"
Converts tracefile to the mongo database that is compatible with Ginan's mongo output for EDA
Determines appropriate filename for GNSS based on the content
Compares two sp3 files and outputs statistics to the terminal window.
How to use:
orbq -i IGS2R03FIN_20191990000_01D_05M_ORB.SP3 TUG0R03FIN_20191990000_01D_05M_ORB.SP3
Result:
PRN R_RMS A_RMS C_RMS dX_RMS dY_RMS dZ_RMS 1D_MEAN 3D_RMS
E01 0.01429 0.00629 0.0077 0.01116 0.00894 0.00993 0.01001 0.01741
E02 0.01931 0.01063 0.01009 0.01564 0.01332 0.01288 0.01395 0.02424
E03 0.0207 0.01334 0.00935 0.0139 0.01909 0.01168 0.01489 0.02634
...
=======================================================================
R_RMS A_RMS C_RMS dX_RMS dY_RMS dZ_RMS 1D_MEAN 3D_RMS
AVG 0.01155 0.01536 0.01346 0.01504 0.01443 0.01277 0.01408 0.02485
STD 0.01045 0.02978 0.01767 0.02065 0.02253 0.01809 0.02017 0.03522
RMS 0.01553 0.03334 0.02212 0.02543 0.02663 0.02205 0.02449 0.04292
Compares two clk files and outputs statistics to the terminal window. How to use:
clkq -i IGS2R03FIN_20191990000_01D_30S_CLK.CLK COD0R03FIN_20191990000_01D_30S_CLK.CLK
Result:
INFO:root:filtering using 10.00 hard cutoff on the detrended data. [99.99% data left]
INFO:root:filtering using 3.00*sigma cut on the detrended data. [99.99% data left]
INFO:root:clkq
CODE AVG STD RMS
E01 3.6127 0.0469 3.613
E02 3.5929 0.0469 3.5932
E03 3.6221 0.042 3.6224
...
==============================
GNSS AVG STD RMS
E 3.6075 0.0595 3.608
G -0.0621 0.0663 0.0909
R 6.1658 0.3282 6.1746
You are able to provide a regex to reject satellites, e.g.:
clkq -i IGS2R03FIN_20191990000_01D_30S_CLK.CLK COD0R03FIN_20191990000_01D_30S_CLK.CLK --reject "G18"
Result:
NFO:root:Excluding satellites based on regex expression: 'G18'
INFO:root:Removed the following satellites from first file: '['G18']'
INFO:root:Removed the following satellites from second file: '['G18']'
INFO:root:filtering using 10.00 hard cutoff on the detrended data. [100.00% data left]
INFO:root:filtering using 3.00*sigma cut on the detrended data. [100.00% data left]
INFO:root:clkq
CODE AVG STD RMS
E01 3.6127 0.0469 3.613
E02 3.5929 0.0469 3.5932
...
And you are also able to provide a normalisation parameter to choose how the data is normalised (daily, epoch), e.g.:
clkq -i IGS2R03FIN_20191990000_01D_30S_CLK.CLK COD0R03FIN_20191990000_01D_30S_CLK.CLK --norm "daily"
Result:
INFO:root::_clk_compare:using ['daily'] clk normalization
INFO:root:---removing common mode from clk 1---
INFO:root:Using daily offsets for common mode removal
INFO:root:---removing common mode from clk 2---
INFO:root:Using daily offsets for common mode removal
INFO:root:filtering using 10.00 hard cutoff on the detrended data. [99.99% data left]
INFO:root:filtering using 3.00*sigma cut on the detrended data. [99.99% data left]
INFO:root:clkq
CODE AVG STD RMS
E01 -0.1216 0.0469 0.1303
E02 -0.1214 0.0469 0.1302
E03 -0.1216 0.042 0.1286
...
These can also be stacked:
clkq -i IGS2R03FIN_20191990000_01D_30S_CLK.CLK COD0R03FIN_20191990000_01D_30S_CLK.CLK --norm "daily" --norm "epoch"
Result:
INFO:root::_clk_compare:using ['daily', 'epoch'] clk normalization
INFO:root:---removing common mode from clk 1---
INFO:root:Using daily offsets for common mode removal
INFO:root:Using epoch normalization (mean gnss) offsets for common mode removal
INFO:root:---removing common mode from clk 2---
INFO:root:Using daily offsets for common mode removal
INFO:root:Using epoch normalization (mean gnss) offsets for common mode removal
INFO:root:filtering using 10.00 hard cutoff on the detrended data. [99.99% data left]
INFO:root:filtering using 3.00*sigma cut on the detrended data. [99.99% data left]
INFO:root:clkq
CODE AVG STD RMS
E01 -0.0 0.0149 0.0149
E02 -0.0 0.0191 0.0191
E03 -0.0 0.0076 0.0076
...
Combination of with a frame file projected to a midday of a date of interest Usage examples:
- Daily comnination with frame_of_day centered at midday
from gnssanalysis import gn_combi
daily_comb_neq = gn_combi.addneq(snx_filelist=_glob.glob('/data/cddis/2160/[!sio][!mig]*0.snx.Z'),frame_of_day=frame_of_day)
- Weekly combination with frame_of_day centered at week's center:
weekly_comb_neq = gn_combi.addneq(snx_filelist=_glob.glob('/data/cddis/2160/[!sio][!mig]*.snx.Z'),frame_of_day=frame_of_day)
The frame of day could be generated using a respective function from gn_frame
module:
from gnssanalysis import gn_frame, gn_datetime
frame_datetime = gn_datetime.gpsweeksec2datetime(2160,43200)
frame_of_day = gn_frame.get_frame_of_day(date_or_j2000=frame_datetime, itrf_path_or_df = '/data/cddis/itrf2014/ITRF2014-IGS-TRF.SNX.gz',discon_path_or_df='/data/cddis/itrf2014/ITRF2014-soln-gnss.snx',psd_path_or_df='/data/cddis/itrf2014/ITRF2014-psd-gnss.snx')