prx reads a RINEX 3.05 observation file and outputs a CSV file of the following format
# {"approximate_receiver_ecef_position_m": [4627852.5264, 119640.514, 4372994.8358], "input_files": [{"name": "TLSE00FRA_R_20240011200_15M_01S_MO.rnx", "murmur3_hash": "bbc2db05fb5d0384ac26e9b9c0236f69"}, {"name": "BRDC00IGS_R_20240010000_01D_MN.rnx", "murmur3_hash": "f7df97043856eb0e57b8e1589c061043"}], "prx_git_commit_id": "ee48104e3d96098a7dc945f7af8ae8484fcd232a", "processing_start_time": "2024-10-02 17:21:08.7285603", "processing_time": "0 days 00:00:19.708024"}
time_of_reception_in_receiver_time,sat_clock_offset_m,sat_clock_drift_mps,sat_pos_x_m,sat_pos_y_m,sat_pos_z_m,sat_vel_x_mps,sat_vel_y_mps,sat_vel_z_mps,ephemeris_hash,frequency_slot,relativistic_clock_effect_m,sagnac_effect_m,tropo_delay_m,sat_code_bias_m,carrier_frequency_hz,iono_delay_m,sat_elevation_deg,sat_azimuth_deg,rnx_obs_identifier,C_obs_m,D_obs_hz,L_obs_cycles,LLI,S_obs_dBHz,constellation,prn
2024-01-01 12:00:00.000000,57153.224736,-0.018397,21838606.222023,36012172.441686,-1479022.231769,-3.184508,4.675864,-0.920381,2245131395643944762,1.000000,-0.668490,-39.902407,10.053189,0.149896,1561098000.000000,-15.044328,13.082414,115.281248,2I,40176280.391000,-14.730000,209208378.723000,,35.400000,C,05
2024-01-01 12:00:00.000000,57153.224736,-0.018397,21838606.222023,36012172.441686,-1479022.231769,-3.184508,4.675864,-0.920381,2245131395643944762,1.000000,-0.668490,-39.902407,10.053189,0.000000,1268520000.000000,-22.784446,13.082414,115.281248,6I,40176273.273000,-11.969000,169998932.830000,,36.900000,C,05
2024-01-01 12:00:00.000000,57153.224736,-0.018397,21838606.222023,36012172.441686,-1479022.231769,-3.184508,4.675864,-0.920381,2245131395643944762,1.000000,-0.668490,-39.902407,10.053189,-2.728111,1207140000.000000,-25.160417,13.082414,115.281248,7I,40176276.934000,-11.391000,161773185.396000,,38.400000,C,05
...
In addition to the observation values extracted from the RINEX file, the rows of the CSV contain the corresponding s 8000 atellite position and speed at the time of transmission as well as relativistic effects, broadcast group delays, broadcast ionosphere delays etc.
From the prx repository root, run
uv sync
uv run python src/prx/main.py --observation_file_path <path_to_rinex_file>
You might have to add <path to prx root>/src/prx to your PYTHONPATH environment variable if you run
into import errors.
We use uv to make sure every developer and user of prx runs the same python version
and the same set of dependencies such as numpy and pandas.
To install uv see https://docs.astral.sh/uv/getting-started/installation/
To create the virtual environment, run
uv sync in the prx repository root. This has to be run every time pyproject.toml or
uv.lock change (for example, after a git pull) to keep your local virtual environment up to date.
Run source .venv/bin/activate to activate the virtual environment in a terminal.
Under File -> Settings -> Project -> Python Interpreter add .venv/bin/python as python interpreter.
Let's say you wrote some code that uses import new_package. To have new package added (what you otherwise
would use pip or some other package manager for), run
uv add new_package
uv will resolve dependencies - i.e. figure out which is the latest version of new_package that is compatible with
our virtual environment - and add it to pyproject.toml and uv.lock.
Run uv run pytest in the prx repository root to run all tests.
Run uv run pytest -x to stop after the first test that fails.
Run uv run pytest -k "my_test" to run a specific test
Run uv run pytest ---durations=10 to run all tests and have pytest list the 10 longest running tests.
We use https://google.github.io/styleguide/pyguide.html as our python style guide.
To find bottlenecks in a specific function, decorate it with @profile, making sure that the module also does from line_profiler import profile.
Then run for instance
LINE_PROFILE=1 # Linux Terminal
# $env:LINE_PROFILE=1 # Windows PowerShell
# set LINE_PROFILE=1 # Windows CMD
uv run kernprof -lb src/prx/test/benchmark.py
uv run python -m line_profiler -rmt --unit 1e-3 --sort "benchmark.py.lprof"
which will print how much time python spent on each line:
Total time: 9.20136 s
File: /Users/janbolting/repositories/prx/src/prx/main.py
Function: build_records at line 239
Line # Hits Time Per Hit % Time Line Contents
==============================================================
239 @prx.util.timeit
240 @profile
241 def build_records(
242 rinex_3_obs_file,
243 rinex_3_ephemerides_files,
244 approximate_receiver_ecef_position_m,
245 ):
246 2 88.2 44.1 1.0 warm_up_parser_cache([rinex_3_obs_file] + rinex_3_ephemerides_files)
247 4 0.0 0.0 0.0 approximate_receiver_ecef_position_m = np.array(
248 2 0.0 0.0 0.0 approximate_receiver_ecef_position_m
249 )
250 2 1.0 0.5 0.0 check_assumptions(rinex_3_obs_file)
251 2 55.7 27.8 0.6 flat_obs = util.parse_rinex_obs_file(rinex_3_obs_file)
...
Please open a PR adding a unit test that reproduces the issue, ideally with an input that is as small as possible to keep our tests lean and fast. No need to be a python developer: If you don't know how to write the test, a PR with a comment describing what you envision the test to do will do - then simply tag a few people on the PR you think can help to get the actual implementation going.
If you don't know how to open a PR, feel free to open an issue instead, describing how to reproduce the unexpected behaviour you are seeing and uploading or linking to the data needed.
Please open a PR adding with a unit test that tests the new feature. No need to be a python developer: If you don't know how to implement the feature or write the test, a PR with a comment describing what the test will do can be enough - then tag a few people you think can help to get the actual implementation going.
If you don't know how to open a PR, feel free to open an issue instead.
When processing your own GNSS data, your receiver may provide you both the RINEX OBS and NAV files. If you want to
specifically use your own RINEX NAV file, it shall be in the same folder as your OBS file and follow the RINEX 3 file
naming convention. This can be achieved by passing your RINEX NAV file in gfzrnx:
.\src\prx\tools\gfzrnx\gfzrnx_217_win64.exe -finp <RINEX_OBS_filepath> -fout ::RX3:: (on Windows, or use the adequate
gfzrnxbinary)
See the Rinex 3.05 spec, page 5, for a list of most acronyms used in the code. Those not covered by the RINEX spec are listed below.
| Acronym | Long Form |
|---|---|
| PRX | Preprocessed Rinex |