LEOCraft - A Flow Level LEO Network Simulator

A modular and extensible LEO network simulation platform to explore traffic flow, throughput, stretch/latency, and coverage. LEOCraft operates at flow level to evaluate a given LEO constellation quickly.

Table of Contents

• Getting Started
  • Setup the Simulation Environment
  • Get the Academic License for Gurobi Optimizer
  • Test the Simulation Environment
  • Unit Tests
  • Simulate with LEOCraft
    • Create LEO Constellation
    • Measure Throughput
    • Measure Stretch/Latency
    • Measure Coverage
  • Visualize with LEOCraft
    • Visualize in 2D
    • Visualize in 3D
  • Examples
    • LEO Constellation Visulization Examples
    • LEO Constellation Simulation Examples
• Extend LEOCraft
  • Explore the API Docs
  • Understand Traffic Modeling
• License
• Cite This Work
• Artifact Evaluation
• Credits

Getting Started

Setup the Simulation Environment

Clone the LEOCraft repository.

git clone https://github.com/suvambasak/LEOCraft.git

Change the directory.

cd LEOCraft

Create a conda environment from environment.yml or environment_macOS.yml.

conda env create -f tools/environment.yml

or

conda env create -f tools/environment_macOS.yml

Activate leocraft environment.

conda activate leocraft

Get the Academic License for Gurobi Optimizer

Since LEOCraft is a flow-level simulator, it uses Gurobi Optimizer to solve the linear program of throughput maximization. To set Gurobi Optimizer:

• Extract the license tools licensetools12.0.1_linux64.tar.gz or download the license tools based on your platform to the binary grbgetkey.

• Sign up and generate free Named-User Academic license.

• To save the license on your system, execute the following.

./grbgetkey <LICENSE_KEY>

Enter and store the gurobi.lic file in the home directory.

info  : grbgetkey version 12.0.1, build v12.0.1rc0
info  : Platform is linux64 (linux) - "Linux Mint 22.1"
info  : Contacting Gurobi license server...
info  : License file for license ID XXXXXXX was successfully retrieved
info  : License expires at the end of the day on XXXX-XX-XX 
info  : Saving license file...

In which directory would you like to store the Gurobi license file?
[hit Enter to store it in /home/suvam]:

Test the Simulation Environment

Export the LEOCraft project path in the shell, which will allow seamless execution of scripts.

export PYTHONPATH=$(pwd)

Or, for instance, if the repository is cloned at /mnt/Storage/Projects/LEOCraft

export PYTHONPATH="/mnt/Storage/Projects/LEOCraft"

To validate the simulation environment setup, execute example_starlink.py, which simulates the Starlink multi-shell LEO constellation.

python examples/example_starlink.py

The LEOCraft setup is successful if you see something like this.

[LEOConstellation] Building ground stations...
[LEOConstellation] Building shells...
[LEOConstellation] Building ground to satellite links...                
[LEOConstellation] GSLs generated in: 0.08m                             
[LEOConstellation] Adding satellites into network graph...
[LEOConstellation] Adding ISLs into network graph...
[LEOConstellation] Routes generated in: 2.2m                            

[Throughput] Building throughput...
[Throughput] Computing throughput...                                    
 LP formation...Set parameter Username
Academic license - for non-commercial use only - expires XXXX-XX-XX
[Throughput] Optimized in: 0.11m                                        
[Throughput] Throughput:    6144.55 Gbps
[Throughput] Total accommodated flow:   30.562 %
[Throughput] NS path selection: 5.929 %
[Throughput] EW path selection: 3.848 %
[Throughput] NESW path selection:   5.325 %
[Throughput] HG path selection: 2.601 %
[Throughput] LG path selection: 7.466 %

[Coverage] Building coverage...
[Coverage] Computing coverage...
[Coverage] Out of coverage GS:  0
[Coverage] GS coverage metric:  148.79394673481124

[Stretch] Building stretch...
[Stretch] Computing stretch...                                          
[Stretch] NS stretch:   1.357
[Stretch] NS hop count: 7.0
[Stretch] EW stretch:   1.504
[Stretch] EW hop count: 8.0
[Stretch] NESW stretch: 1.253
[Stretch] NESW hop count:   6.0
[Stretch] LG stretch:   1.434
[Stretch] LG hop count: 3.5
[Stretch] HG stretch:   1.181
[Stretch] HG hop count: 11.0
Total simulation time: 2.44m

Unit Tests

To run the unit tests, follow the instructions in tests.

Simulate with LEOCraft

Here is the quick overview simulating an LEO constellation network of Starlink.

Create LEO Constellation

First, import the LEOCraft modules.

from LEOCraft.attenuation.fspl import FSPL
from LEOCraft.constellations.LEO_constellation import LEOConstellation
from LEOCraft.dataset import GroundStationAtCities
from LEOCraft.satellite_topology.plus_grid_shell import PlusGridShell
from LEOCraft.user_terminals.ground_station import GroundStation

Create an instance of the atmospherics path loss model.

loss_model = FSPL(
    28.5*1000000000,    # Frequency in Hz
    98.4,               # Tx power dBm
    0.5*1000000000,     # Bandwidth Hz
    13.6                # G/T ratio
)
loss_model.set_Tx_antenna_gain(gain_dB=34.5)

To create a LEO constellation of Starlink, create an object of LEO constellation, add ground stations, and then add the first shell of Starlink of +Grid topology.

leo_con = LEOConstellation('Starlink')
leo_con.add_ground_stations(
 GroundStation(
 GroundStationAtCities.TOP_100
 )
)

leo_con.add_shells(
 PlusGridShell(
        id=0,
        orbits=72,
        sat_per_orbit=22,
        altitude_m=550000.0,
        inclination_degree=53.0,
        angle_of_elevation_degree=25.0,
        phase_offset=50.0
 )
)

leo_con.set_time(second=3)  # Time passed after epoch
leo_con.set_loss_model(loss_model)
leo_con.build()
leo_con.create_network_graph()

leo_con.generate_routes()

You will see this on the execution.

[LEOConstellation] Building ground stations...
[LEOConstellation] Building shells...
[LEOConstellation] Building ground to satellite links...                
[LEOConstellation] GSLs generated in: 0.11m                             
[LEOConstellation] Adding satellites into network graph...
[LEOConstellation] Adding ISLs into network graph...
[LEOConstellation] Routes generated in: 2.83m 

Measure throughput

To measure the throughput of the above Starlink constellation, import the following modules.

from LEOCraft.dataset import InternetTrafficAcrossCities
from LEOCraft.performance.basic.throughput import Throughput

Create the instance of throughput, then add the instance of Starlink constellation and traffic matrics across the ground stations to compute the throughput.

th = Throughput(
 leo_con,
 InternetTrafficAcrossCities.ONLY_POP_100
)
th.build()
th.compute()

The output will be as follows.

[Throughput] Building throughput...
[Throughput] Computing throughput...                                    
 LP formation...Set parameter Username
Set parameter LicenseID to value XXXXXXX
Academic license - for non-commercial use only - expires 20XX-XX-XX
[Throughput] Optimized in: 0.02m                                        
[Throughput] Throughput:        2820.032 Gbps
[Throughput] Total accommodated flow:   15.209 %
[Throughput] NS path selection: 1.77 %
[Throughput] EW path selection: 1.455 %
[Throughput] NESW path selection:       1.916 %
[Throughput] HG path selection: 0.999 %
[Throughput] LG path selection: 6.945 %

Measure Stretch/Latency

To measure stretch, import the following module.

from LEOCraft.performance.basic.stretch import Stretch

Create an instance of stretch and add the instance of Starlink constellation to compute stretch.

sth = Stretch(leo_con)
sth.build()
sth.compute()

The output will be as follows.

[Stretch] Building stretch...
[Stretch] Computing stretch...                                          
[Stretch] NS stretch:   1.868
[Stretch] NS hop count: 8.0
[Stretch] EW stretch:   1.668
[Stretch] EW hop count: 8.0
[Stretch] NESW stretch: 1.285
[Stretch] NESW hop count:       6.0
[Stretch] LG stretch:   1.479
[Stretch] LG hop count: 4.0
[Stretch] HG stretch:   1.263
[Stretch] HG hop count: 12.0

Measure Coverage

To measure the coverage of the constellation, import the following module.

from LEOCraft.performance.basic.stretch import Stretch

Create an instance of coverage and add the instance of Starlink constellation to compute coverage.

cov = Coverage(leo_con)
cov.build()
cov.compute()

The output will be as follows.

[Coverage] Building coverage...
[Coverage] Computing coverage...
[Coverage] Out of coverage GS:  0
[Coverage] GS coverage metric:  111.9799753395037

Visualize with LEOCraft

The LEOCraft provides excellent 2D and 3D interactive visualization capabilities to intuitively understand the LEO network topology and the transition of topology with LEO dynamics.

    

Visualize in 2D

To visualize Starlink constellation in 2D, use the SatView2D module. Add the components (satellites, ground stations, routes, satellite coverage, etc.) for rendering, as shown below.

from LEOCraft.visuals.sat_view_2D import SatView2D

view = SatView2D(leo_con, default_zoom=2.0)

view.add_all_satellites()
view.add_coverages('S0-0', 'S0-1', 'S0-40', 'S0-31', 'S0-30')
view.add_routes('G-0_G-1', 'G-1_G-2', 'G-2_G-3', 'G-30_G-33', k=1)

view.build()

view.export_html('docs/html/Starlink_2D.html')

The output will be as follows.

[SatView2D] Building view 2D...  
[SatView2D] Exporting HTML...  

The interactive 2D visualization Starlink_2D.html.

Visualize in 3D

To visualize the Starlink constellation in 3D, use the SatView3D module. Add the add the components (satellites, ground stations, routes, satellite coverage, etc) for rendering as shown in the following.

from LEOCraft.visuals.sat_view_3D import SatView3D

view = SatView3D(leo_con)

view.add_all_satellites()
view.add_routes('G-0_G-1', 'G-1_G-2', 'G-2_G-3', 'G-30_G-33', k=1)

view.build()

view.export_html('docs/html/Starlink_3D.html')

The output will be as follows.

[SatView3D] Building raw view 3D...  
[SatView3D] Exporting HTML...  

The interactive 2D visualization Starlink_3D.html.

To visualize the Starlink constellation in 3D but without maps and satellites at elevation, use the SatRawView3D module. Add the components (satellites, ground stations, routes, satellite coverage, etc.) for rendering, as shown below.

from LEOCraft.visuals.sat_raw_view_3D import SatRawView3D

view = SatRawView3D(
 leo_con,

    lat=16.80528,
    long=96.15611,
    elevation_m=550000
)


view.add_all_satellites()
view.add_routes('G-0_G-1', k=1)

view.build()

view.export_html('docs/html/Starlink_RAW_3D.html')

The output will be as follows.

[SatRawView3D] Building raw view 3D...  
[SatRawView3D] Exporting HTML... 

The interactive 2D visualization Starlink_RAW_3D.html.

    

Examples

LEO Constellation Visulization Examples

Following are some examples of interactive LEO network visualisation. A few videos of LEO dynamics are also available at LEOCraft Playlist.

• LEO consetalltion design

  • Starlink multi shell design
  • Kuiper multi shell design
  • Telesat multi shell design

• Ground station locations

  • At 100 most populous city across the world
  • At 1000 most populous city across the world
  • At all country capitals

• Starlink coverage change with altitude

  • At altitude 300 km
  • At altitude 600 km

• Starlink inclination change

  • At inclination 5 degree
  • At inclination 40 degree
  • At inclination 90 degree

• Starlink high geodesic distance path diversity

  • When o=144, n=11, p=50
  • When o=144, n=11, p=0
  • When o=48, n=33, p=50

• Proposed inter shell ISL connectivity

  • Between two shells
  • Between three shells

• ISLs usage visualization

  • With 100 most populous city across the world
  • With all country capitals

• ISL topological structures tweak

  • orbits=72, sat_per_orbit=22, phase_offset=50.0
  • orbits=22, sat_per_orbit=72, phase_offset=50.0
  • orbits=44, sat_per_orbit=36, phase_offset=50.0
  • orbits=44, sat_per_orbit=36, phase_offset=0.0

• Route change with inclination and elevation angle

  • Inclination 90 degree
  • Inclination 50 degree
  • Inclination 30 degree
  • Elevation angle 10 degree
  • Elevation angle 30 degree
  • Elevation angle 50 degree

• Number of hops in route reduces with higher altitude

  • Altitude 550 km
  • Altitude 1000 km

• Route categories based on orientation

  • North East to South West
  • East to West
  • North to South
  • High Geodesic Distance

• Others

  • Satellite ground track
  • 2D view of route, GS, GSLs satellites, satellite coverage
  • 3D view of route, GS, GSLs satellites
  • 3D RAW view of route, GS, GSLs satellites
  • Coverage cone change with value of elevation angle

LEO Constellation Simulation Examples

A handful of examples of Python scripts for programming the LEO network simulation and generating interactive visuals are available in examples.

• Simulate Starlink multi shell constellation with ground stations
• Simulate a batch of LEO constellation with ground stations
• Simulate Starlink multi shell constellation with flights
• Simulate a batch of LEO constellation with flights
• Simulate LEO constellation with inter-shell ISLs
• Visualize LEO constellation in 2D
• Visualize LEO constellation in RAW 3D
• Visualize LEO constellation in 3D
• Render evolution LEO network frame by frame
• Archive of proposed shell designs

Extend LEOCraft

LEOCraft is designed with modularity and extensibility at its core. Whether you're exploring custom satellite constellations, novel inter-satellite routing schemes, new traffic models, or advanced performance metrics, LEOCraft provides the scaffolding to implement and test your ideas quickly and rigorously.

Where to Start

Explore the API Docs

Dive into the API documentation to understand the architecture and core components of LEOCraft—from constellation modeling and link computation to routing and performance evaluation.

Understand Traffic Modeling

Check out the traffic dataset source to learn how Internet demand between cities is modeled and how you can plug in your own demand matrices or traffic sources.

License

This work is licensed under the MIT License.

Cite This Work

Plain text

Basak, S., Pal, A. and Bhattacherjee, D., 2025. {LEOCraft}: Towards Designing Performant {LEO} Networks. In 2025 USENIX Annual Technical Conference (USENIX ATC 25) (pp. 789-813).

BibTeX

@inproceedings{LEOCraft,
  title={$\{$LEOCraft$\}$: Towards Designing Performant $\{$LEO$\}$ Networks},
  author={Basak, Suvam and Pal, Amitangshu and Bhattacherjee, Debopam},
  booktitle={2025 USENIX Annual Technical Conference (USENIX ATC 25)},
  pages={789--813},
  year={2025}
}

Artifact Evaluation

For executing the experiments and regenerating the figures in the paper, straightforward steps are given in ARTIFACT EVALUATION.

Credits

Contributors whose support and expertise have been invaluable in the development of LEOCraft:

• Amitangshu Pal
• Debopam Bhattacherjee