GNSSpy v3

GNSSpy v3.0.1 is an open-source Python library for RINEX conversion, multi-GNSS data handling, orbit analysis, atmospheric corrections, positioning support, quality diagnostics and interactive visualisation.

Version 3 reorganises GNSSpy around scientific functions rather than the former project layout. Data access, GNSS file readers, RINEX conversion, broadcast and precise orbit tools, atmospheric models, positioning routines, quality-control functions, visualisation and executable workflows are now exposed through separate package namespaces.

Main capabilities include:

RINEX 2 ↔ RINEX 3 observation-file conversion
Authenticated GNSS data download from NASA CDDIS
RINEX observation and navigation reading
SP3, CLK and IONEX product handling
Broadcast-orbit computation and SP3 interpolation
Ionospheric and tropospheric corrections
Standard point positioning and observation–orbit matching utilities
SNR, multipath and satellite-visibility diagnostics
Interactive Plotly visualisation
Single-epoch and full-day BRDC–SP3 orbit-comparison workflows

Current orbit-comparison scope: the packaged BRDC–SP3 workflows currently compare GPS, Galileo and BeiDou satellites. Other constellations remain available in the broader file-reading, data-handling and visualisation layers where suitable observations and orbit products are available.

Installation

GNSSpy v3 requires Python 3.10 or later.

Recommended installation from GitHub

git clone https://github.com/GNSSpy-Project/gnsspy.git
cd gnsspy

python3 -m venv .venv
source .venv/bin/activate          # Windows: .venv\Scripts\activate

python -m pip install --upgrade pip
python -m pip install -e ".[all]"

The all dependency group installs the packages needed by the visualisation, product-processing and orbit-comparison workflows. A smaller core installation is also available:

python -m pip install -e .

Optional groups can be installed separately:

python -m pip install -e ".[visualization]"
python -m pip install -e ".[products]"
python -m pip install -e ".[workflows]"

Hatanaka decompression

GNSSpy includes the RNXCMP CRX2RNX executables used to decompress Hatanaka observation files (.crx and .d). The library searches for a compatible executable in the following order:

the system PATH;
the installed package under gnsspy/bin/;
the source-tree bin/ directory.

The bundled files cover Linux and Windows. On macOS, install a compatible CRX2RNX executable and place it on the system PATH.

The standalone RINEX 2 ↔ 3 converter handles ordinary observation files and their .gz or .Z compression. A Hatanaka file must first be expanded with CRX2RNX or gnsspy.io.manipulate.crx2rnx().

Quick start

Run the central interface from the project root:

python -m gnsspy

After installation, the equivalent command is:

gnsspy

The main menu is:

======================================================================
                 GNSSPY - CENTRAL MANAGEMENT INTERFACE
======================================================================

  1. Download GNSS Data
  2. Visualize GNSS Data (Skyplots, SNR, Groundtrack)
  3. Orbit Comparison (Single Epoch)
  4. Orbit Comparison (Full Day Time Series)
  5. RINEX Converter (2 <-> 3)
  6. Exit

The installed package also provides direct commands:

gnsspy-download
gnsspy-visualize
gnsspy-convert-rinex

The orbit-comparison workflows are currently launched through the central gnsspy menu or imported from gnsspy.workflows.

Command-line interfaces

1) Download GNSS data

Run either:

gnsspy-download

or select Option 1 from the main menu.

The interactive downloader guides you through:

NASA Earthdata login — credentials are checked before the download begins. A credentials.txt file in the current working directory can be used to avoid re-entering them.
Date selection — one day or an inclusive date range. Common formats such as YYYY-MM-DD, DD-MM-YYYY, slash-separated and dot-separated dates are accepted.
Station selection — comma-separated station codes such as MATE, ANKR, ISTA. Use BRDC when the task requires only a global merged broadcast file.
RINEX version — version 2 or 3. The requested version is tried first; the downloader can also test the alternative archive format when availability differs by period.
File selection — any combination of:
- observation files;
- station navigation files;
- merged BRDC navigation files;
- SP3 precise orbits and CLK clock products;
- IONEX ionosphere products.
Analysis centre for SP3 and CLK products:
- CODE — Center for Orbit Determination in Europe;
- GFZ — GFZ German Research Centre for Geosciences;
- IGS — International GNSS Service combination;
- WUM — Wuhan University multi-GNSS products;
- MIT — Massachusetts Institute of Technology products.
Output directory — use the configured data/ directory or provide another location.

For SP3 interpolation, GNSSpy downloads the day before, the target day or range, and the day after. The selected centre is used in strict mode for the whole interval; the downloader does not silently mix products from different centres.

Files are organised into product-specific subdirectories such as:

<output>/observation/
<output>/navigation/
<output>/brdc/
<output>/sp3/
<output>/clk/
<output>/ionosphere/

A per-file success or failure message is printed during the run, followed by a summary report.

2) Visualise GNSS data

Run either:

gnsspy-visualize

or select Option 2 from the main menu.

The visualiser can use an existing data directory or trigger the downloader. It produces interactive HTML output with Plotly.

#	Plot	Required input	Description
1	Skyplot	Observation + orbit	Polar view of satellite tracks, optionally coloured by SNR
2	Azimuth–Elevation	Observation + orbit	Azimuth and elevation through time
3	Elevation Time Series	Observation + orbit	Elevation history for the selected satellites
4	SNR Time Series	Observation	Signal-to-noise ratio through time
5	Bandplot / Visibility	Observation	Availability of observation bands and signals by satellite
6	Groundtrack	Orbit	Satellite ground tracks
7	All plots	Observation + orbit	Generates every available plot

The system selector accepts:

G = GPS
R = GLONASS
E = Galileo
C = BeiDou
I = IRNSS/NavIC
J = QZSS
S = SBAS

For GPS, GLONASS, Galileo and BeiDou visualisation, orbit-dependent plots normally use SP3 products. For IRNSS/NavIC and QZSS, the visualiser switches to broadcast navigation because these constellations are not generally included in the selected standard SP3 products.

Compressed observation files are expanded as needed. Hatanaka .crx or .d inputs are passed to CRX2RNX before reading.

HTML files are written to:

<output>/analyses/

and opened in the default browser when processing finishes.

3) Orbit comparison — single epoch

Select Option 3 from the main menu.

This workflow compares broadcast and precise satellite coordinates at one UTC epoch:

BRDC coordinates are propagated from the selected broadcast ephemeris records.
SP3 coordinates are obtained through GNSSpy polynomial interpolation, using adjacent-day products around the target day.

The workflow:

uses an existing directory or downloads navigation, BRDC, SP3 and CLK files;
discovers the available dates from the navigation files;
asks for a UTC time;

accepts flexible satellite filters such as:

all
gps
galileo
beidou
gps2-23-24
galileo01:10
beidou12-15
g05,e12,c07
gps,galileo

calculates ECEF coordinate differences dX, dY, dZ and the three-dimensional difference;
rejects non-finite results and differences above the current 100 m quality ceiling.

The current workflow evaluates GPS, Galileo and BeiDou broadcast ephemerides. Results are exported to:

<output>/data/georinex_portable_YYYYMMDD_HHMMSS.xlsx

The spreadsheet contains BRDC and SP3 ECEF coordinates, component differences and the total three-dimensional difference for each retained satellite.

4) Orbit comparison — full-day time series

Select Option 4 from the main menu.

This workflow extends the BRDC–SP3 comparison over a complete UTC day at a default 30-second interval. SP3 coordinates are interpolated with the workflow's 16th-degree polynomial setting.

The same flexible satellite filters used by the single-epoch workflow are available. The current comparison covers GPS, Galileo and BeiDou.

Before export, the workflow applies several quality checks:

unhealthy broadcast ephemeris records are excluded;
epochs outside the current broadcast-ephemeris validity window are removed;
coordinate or three-dimensional differences above 100 m are rejected;
an additional per-satellite median-absolute-deviation filter removes isolated residual outliers.

Outputs include:

one Excel file containing every retained satellite–epoch result;
per-system three-dimensional RMS bar charts;
system-wide satellite overlays for dX, dY and dZ;
per-satellite BRDC and SP3 coordinate comparisons;
per-satellite dX, dY, dZ and three-dimensional difference time series.

Files are saved under:

<output>/data/timeseries_YYYY-MM-DD.xlsx
<output>/data/figures/

Figures are written as 300 DPI PNG files.

5) Convert RINEX observation files

Run either:

gnsspy-convert-rinex

or select Option 5 from the main menu.

The interactive converter supports both directions:

RINEX 2 -> RINEX 3.04
RINEX 3 -> RINEX 2.11

A source file can be supplied by:

pasting or dragging a local file path;
selecting a file discovered inside an existing directory;
downloading a plain RINEX 2 observation file from CDDIS for conversion to RINEX 3.

The converter engine reads ordinary observation files, gzip-compressed files and Unix-compress .Z files. Hatanaka .d and .crx files must be decompressed first.

When writing RINEX 2.11, the interactive tool retains the standard G/R/E/S group or the smaller G/R group. BeiDou, QZSS and IRNSS/NavIC observations cannot be represented by this RINEX 2.11 output mode and are therefore omitted.

Non-interactive converter

The converter also has an argument-based interface:

python -m gnsspy.io.rinex.converter \
    input_file.rnx \
    output_file.25o \
    --target 2.11 \
    --keep GRES

For the opposite direction:

python -m gnsspy.io.rinex.converter \
    input_file.25o \
    output_file.rnx \
    --target 3.04

Programmatic use is available through:

from gnsspy.io.rinex.converter import convert_file

convert_file(
    "input_file.25o",
    "output_file.rnx",
    target_version=3.04,
)

Python API

GNSSpy v3 can be used as a conventional Python library without the interactive interfaces.

Main namespaces

import gnsspy
import gnsspy.data_access
import gnsspy.io
import gnsspy.orbit
import gnsspy.atmosphere
import gnsspy.positioning
import gnsspy.quality
import gnsspy.geodesy
import gnsspy.visualization
import gnsspy.workflows

print(gnsspy.__version__)

Read GNSS files

from gnsspy.io.rinex.observation import read_obsFile
from gnsspy.io.rinex.navigation import read_navFile
from gnsspy.io.products.sp3 import read_sp3File
from gnsspy.io.products.clk import read_clockFile
from gnsspy.io.products.ionex import read_ionFile

observation = read_obsFile("station_file.rnx")
navigation = read_navFile("broadcast_navigation.rnx")
sp3 = read_sp3File("precise_orbit.sp3")
clock = read_clockFile("precise_clock.clk")
ionosphere = read_ionFile("ionosphere.inx")

The historical GNSSpy function names remain available, while the v3 namespaces also expose more explicit aliases such as read_observation_file, read_navigation_file, read_sp3_file, read_clock_file and read_ionex_file.

Download data from Python

from gnsspy.data_access import ObservationDownloader, NavigationDownloader

obs_downloader = ObservationDownloader(
    username="EARTHDATA_USERNAME",
    password="EARTHDATA_PASSWORD",
    output_dir="data",
)

obs_downloader.download_single(
    station="MATE",
    date="2025-11-02",
    rinex_version=3,
)

product_downloader = NavigationDownloader(
    username="EARTHDATA_USERNAME",
    password="EARTHDATA_PASSWORD",
    output_dir="data",
)

product_downloader.download_brdc_only(
    date="2025-11-02",
    rinex_version=3,
)

Broadcast and precise orbits

import datetime

from gnsspy.orbit.broadcast import calculate_orbit_from_nav
from gnsspy.orbit.precise import sp3_interp

broadcast = calculate_orbit_from_nav(
    navigation=navigation,
    t_start=datetime.datetime(2025, 11, 2, 0, 0, 0),
    t_end=datetime.datetime(2025, 11, 2, 23, 59, 59),
    interval=30,
    system_filter="E",
)

precise = sp3_interp(
    epoch=datetime.date(2025, 11, 2),
    interval=30,
    poly_degree=16,
    sp3_product="cod",
    clock_product="cod",
    data_dir="data",
)

sp3_interp() expects SP3 products for the previous, target and following days, together with the target-day clock product, under the corresponding sp3/ and clk/ directories.

Atmosphere, positioning and diagnostics

from gnsspy.atmosphere.troposphere import tropospheric_delay
from gnsspy.atmosphere.ionosphere import ionosphere_interp
from gnsspy.positioning.spp import spp
from gnsspy.quality.multipath import multipath
from gnsspy.quality.snr import standardize_snr
from gnsspy.visualization import (
    skyplot,
    azelplot,
    timelplot,
    bandplot,
    groundtrack,
)

NASA Earthdata account

GNSSpy downloads observation and product files from the NASA CDDIS archive, which requires an Earthdata Login.

Create an account at NASA Earthdata Login.
Ensure that your account is authorised for CDDIS access.
Enter the credentials when the GNSSpy downloader requests them.

To avoid repeated prompts, create credentials.txt in the directory from which the command is run:

username = your_username
password = your_password

The following two-line form is also accepted:

your_username
your_password

The visualisation credential flow can optionally save credentials in the standard ~/.netrc file. On Unix-like systems, GNSSpy writes that file with restricted permissions.

Security: credentials.txt and .netrc contain sensitive information. Do not commit either file. The repository .gitignore excludes credentials.txt, but file permissions and repository history should still be checked before publishing changes.

Project layout

gnsspy/
├── pyproject.toml                   # Build metadata, dependencies and commands
├── requirements.txt                # Core runtime requirements
├── LICENSE                          # MIT licence
├── MANIFEST.in                      # Source-package inclusion rules
├── bin/                             # Source-tree RNXCMP executables and licence
├── docs/
│   ├── USER_MANUAL.md
│   ├── PACKAGE_STRUCTURE.md
│   ├── RELEASE_NOTES.md
│   └── PATCH_NOTES_v3_0_1.md
└── gnsspy/
    ├── __init__.py                  # Version and compatibility-level imports
    ├── __main__.py                  # python -m gnsspy
    ├── atmosphere/                  # Ionospheric and tropospheric corrections
    ├── bin/                         # Packaged CRX2RNX executables and licence
    ├── cli/                         # Main, download, visualisation and converter CLIs
    ├── core/                        # Shared core namespace
    ├── data/                        # Station and reference data
    ├── data_access/                 # Earthdata/CDDIS sessions and downloaders
    ├── geodesy/                     # Coordinate and projection routines
    ├── io/
    │   ├── products/                # SP3, CLK and IONEX readers
    │   └── rinex/                   # Observation/navigation readers and converter
    ├── orbit/                       # Broadcast, precise and comparison utilities
    ├── positioning/                 # SPP, adjustment and observation matching
    ├── quality/                     # SNR, multipath and visibility diagnostics
    ├── utils/                       # Dates, filenames, constants and interpolation
    ├── visualization/               # Plotly-based diagnostic figures
    └── workflows/                   # Single-epoch and daily BRDC–SP3 workflows

Dependencies

Core dependencies

Package	Used for
`numpy`	Numerical arrays and GNSS calculations
`pandas`	Indexed observation, orbit and product data
`requests`	Authenticated CDDIS downloads
`unlzw3`	Unix-compress `.Z` decompression

Optional dependency groups

Group	Packages	Used for
`visualization`	`matplotlib`, `plotly`, `scipy`	Interactive and static figures plus interpolation support
`products`	`scipy`, `georinex`	Precise-product and broadcast-navigation processing
`workflows`	`matplotlib`, `plotly`, `scipy`, `georinex`, `openpyxl`	Complete orbit-comparison workflows and Excel export
`all`	all optional packages above	Full GNSSpy v3 installation

Install a group with, for example:

python -m pip install -e ".[workflows]"

Documentation and release notes

The repository includes:

docs/USER_MANUAL.md — public API and command overview;
docs/PACKAGE_STRUCTURE.md — v3 namespace organisation;
docs/RELEASE_NOTES.md — major-version notes;
docs/PATCH_NOTES_v3_0_1.md — fixes included in v3.0.1.

Version 3.0.1 includes packaging and stability corrections for date parsing, coordinate conversion, navigation-field parsing, multipath processing, CRX2RNX discovery, Earthdata credential handling, Plotly renderer behaviour and optional dependency management.

Citation

If GNSSpy contributes to academic work, please cite the original GNSSpy publication:

Işık, M. S., Özbey, V., Erol, S., & Tarı, E. (2021). GNSSpy: Python Toolkit for GNSS Data. 2021 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), 8550–8553.

When using GNSSpy v3 specifically, also report the software version and repository, for example:

GNSSpy v3.0.1, https://github.com/GNSSpy-Project/gnsspy

For Hatanaka compression and RNXCMP, cite:

Hatanaka, Y. (2008). A Compression Format and Tools for GNSS Observation Data. Bulletin of the Geospatial Information Authority of Japan, 55, 21–30.

License and third-party components

GNSSpy is distributed under the MIT License. See LICENSE.

Third-party components include:

CRX2RNX and crx2rnx.exe — RNXCMP software from the Geospatial Information Authority of Japan. The applicable licence is included at gnsspy/bin/RNX2CMP_LICENSE.txt and bin/RNX2CMP_LICENSE.txt.

Use and redistribution of the bundled RNXCMP components must follow their accompanying licence terms.

Name		Name	Last commit message	Last commit date
Latest commit History 19 Commits
bin		bin
docs		docs
gnsspy		gnsspy
.gitignore		.gitignore
LICENSE		LICENSE
MANIFEST.in		MANIFEST.in
README.md		README.md
pyproject.toml		pyproject.toml
requirements.txt		requirements.txt

Uh oh!

Folders and files

Latest commit

History

Repository files navigation

GNSSpy v3

Table of contents

Installation

Recommended installation from GitHub

Hatanaka decompression

Quick start

Command-line interfaces

1) Download GNSS data

2) Visualise GNSS data

3) Orbit comparison — single epoch

4) Orbit comparison — full-day time series

5) Convert RINEX observation files

Non-interactive converter

Python API

Main namespaces

Read GNSS files

Download data from Python

Broadcast and precise orbits

Atmosphere, positioning and diagnostics

NASA Earthdata account

Project layout

Dependencies

Core dependencies

Optional dependency groups

Documentation and release notes

Citation

License and third-party components

About

Resources

License

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Uh oh!

Contributors

Uh oh!

Languages

Packages