NetSim Astra Q&A

Q1. What is NetSim Astra, and what problem does it solve?

NetSim Astra is a web-based tool for satellite RF coverage planning. A single configuration holds the constellation, the RF parameters, and the analysis settings. One run propagates the constellation with SGP4, computes an ITU-R link budget at every grid point and every time step, renders the coverage heatmap on a 3D globe, produces an access and gap report, and writes per-sample CSV files.

C/N₀, C/N, G/T, E_b/N₀, BER, and power flux density are computed for every sample of every run, without additional configuration. The scope is limited to this workflow: Astra does not perform orbit design, attitude and sensor modelling, waveform generation, or packet-level protocol simulation.

Q2. We already use Ansys STK. What does Astra do differently?

STK is a mission-engineering environment. Orbit mechanics is the reference frame, and a communications study is assembled around it: a Transmitter object is attached to each transmitting asset, a Receiver object to each receiving asset, a CoverageDefinition sets the grid, the Grid Constraint Options are pointed at the receiver, and an Access Constraint Figure of Merit is defined that references that constraint. The result is exported report by report.

In Astra the same study is three steps: select the constellation, set the RF parameters and the region of interest, and run. One pass produces the coverage heatmap, the access and gap report, and the per-sample CSV files, with C/N₀, C/N, G/T, BER, and PFD at every grid point. Both tools can produce communications-quality coverage. In STK it is assembled from objects, constraints, and figures of merit; in Astra it is the default output of one run.

Two modelling differences. First, on clutter: STK documents an ITU-R propagation chain (P.618 rain, P.676 gases, P.840 clouds, scintillation), terrain propagation through TIREM, and an urban propagation model, but does not document a land-cover clutter model, and there is no ITU-R P.2108 implementation in its published model list. Astra classifies every grid point against ESA WorldCover and applies a per-class clutter loss. Second, on regulatory screening: STK exposes flux density, PFD, and spectral PFD as receiver constraints, and does not document an ITU Radio Regulations Article 21 elevation-dependent limit mask. Astra applies that limit mask and reports a PFD margin at every grid point.

Q3. We already use MATLAB and the Satellite Communications Toolbox. What does Astra do differently?

MATLAB is a programming environment. In the coverage workflow, each step is code the analyst writes: the satelliteScenario, the satellites (walkerStar or a TLE file obtained outside MATLAB), a transmitter on each satellite, the region grid (shapefile, projection, point-in-polygon filter, using the Mapping Toolbox), a loop creating a groundStation and receiver at every grid point, the field-of-view and link loop calling sigstrength per link, the p618PropagationLosses call per point, the contour and geoplot code, and the post-processing of accessIntervals tables into coverage and gap statistics.

MathWorks' Coverage Maps for Satellite Constellation example defines its helper functions inside the example script rather than shipping them in the toolbox, among them the link loop (satcoverage), the footprint geometry (satfov), and the contour gridding (contourDataGrid). A live TLE fetch by category, and area coverage, revisit, and response-time statistics, have no toolbox function and are also user code. Adopting the workflow makes these functions part of the team's codebase, to be maintained across MATLAB releases.

The coverage map itself has no GUI path in MATLAB: there is no coverage app, coverage function, or Live Editor task for it in the current release. The Satellite Link Budget Analyzer app analyses explicitly defined point-to-point links, an uplink, a downlink, or a crosslink, rather than a coverage grid, and the satellite scenario viewer displays a scenario a script has already built. In Astra the three configuration tabs write one config.json, the headless CLI reads that same file, and the coverage map is a product feature rather than user code.

Q4. Can we see the capabilities side by side?

The table below compares the three for one workflow: producing and validating an RF coverage map over a region. It does not assess the products outside that workflow. For a mission-design or a waveform-design task, the rows would be different ones. Licensing and packaging are not compared here; those are matters for each vendor to state, and they change between releases.

Q9. Which constellation sources are supported?

Four, and each one works with every analysis mode:

• Synthetic Walker. Walker Delta or Walker Star, with configurable planes, satellites per plane, altitude, and inclination. Up to 100 planes and 360 satellites per plane, at altitudes to 36,000 km, so LEO through GEO.
• Live TLEs. Starlink, OneWeb, GNSS, GEO operators and other CelesTrak or N2YO categories, with a configurable polling interval.
• Single fixed satellite. A stationary satellite at any position, for antenna-pattern testing and deterministic single-timestep runs.
• GEO and IGSO. Manual entry, INSAT and GSAT presets, or live TLEs. GEO adds the multibeam CINR analysis.

Q10. How are satellite positions propagated?

Live satellites are propagated from their TLEs with SGP4, then transformed into ground-relative geometry for elevation, slant range, and off-axis angle. Positions and angles are computed across Earth-Centered Earth-Fixed (ECEF), East-North-Up (ENU), and Local-Vertical-Local-Horizontal (LVLH) antenna frames. Synthetic scenarios step through a configurable simulation window; live scenarios poll in real time. Each step yields a complete link budget at every ground point.

Q11. What exactly is in the link budget?

The chain runs from the satellite EIRP down to received power at the ground terminal:

• EIRP, receive gain, and antenna pattern gain evaluated at the true off-axis angle
• Free space path loss from the slant range
• ITU-R P.618 atmospheric losses: rain fade, gaseous absorption, cloud attenuation, and tropospheric scintillation, with link availability by exceedance conversion
• Land-use clutter loss from the ESA WorldCover class at the point
• Polarization loss, additional losses, and a system margin

Each component is a separate column in the output CSV. Clicking any heatmap point in the browser displays the same breakdown for that point.

Q12. Which antenna pattern models are available?

• Gaussian: a simplified analytical model
• Bessel (3GPP TR 38.811): circular aperture
• ITU-R S.672-4: the FSS reference pattern
• Phased array (URA): beam shaping for 5G NTN, to 64 × 64 elements
• Custom: CSV gain tables over off-boresight and azimuth angles, or Ansys HFSS Far-Field Data (FFD) files, with 2D interpolation for asymmetric beams and measured sidelobes

The View Pattern tool renders any of these patterns before a run, showing the beam shape and the sidelobe structure for both the analytical models and imported patterns.

Q13. How is clutter loss modelled?

Each ground point is classified against ESA WorldCover land-cover data, and a clutter loss is applied for that class. Defaults run from about 0.5 dB over water to about 8 dB in built-up areas, across 11 classes. Every per-class value and the fallback are editable and are saved with the project, so a site-specific clutter model travels with the study.

Q14. Beyond received power, what link-quality metrics are produced?

• C/N₀ and C/N, in dB-Hz and dB, computed from the receiver system noise temperature and the channel bandwidth, and used to determine whether the link closes against a required threshold
• G/T, the receiver figure of merit in dB/K, with 290 K assumed when no system noise temperature is given
• E_b/N₀, from the configured bit rate
• Doppler shift, from the radial velocity between satellite and ground station. A 550 km LEO satellite reaches roughly ±280 kHz at Ku-band, and about ±37 kHz at L-band

Q15. Is bit error rate computed, and for which modulations?

Yes, in the point, coverage, and multibeam analyses. BER is computed from E_b/N₀ and the configured bit rate for BPSK, QPSK, OQPSK, 8PSK, 16QAM, and 64QAM, together with a code-rate based coded-BER estimate and the symbol rate. It is evaluated at every grid point, so the coverage grid holds a BER value per point in addition to received power.

Q16. What does Astra do about power flux density and regulatory limits?

It reports the incident power density at the ground and checks it against a limit mask:

• Free-space PFD (dBW/m²) from the effective EIRP and the slant range
• PFD after atmospheric attenuation
• Spectral PFD normalized to a selectable reference bandwidth
• An ITU Radio Regulations Article 21 style elevation-dependent limit mask, with presets for C-band (3.7–4.2 GHz) and K-band (17.7–19.7 GHz), and custom limits and reference bandwidths
• A PFD margin on every sample, positive when the point is below the selected limit

This is a screening tool for engineering use, and it reports margins against a limit mask you select. It is not a filing, and it is not a substitute for a coordination or regulatory submission.

Q17. Can multibeam payloads and co-channel interference be evaluated?

Yes, for GEO satellites entered manually. Astra evaluates multibeam payloads with frequency reuse from FR1 to FR4 and produces a CINR heatmap across the service area, which is where co-channel interference between beams becomes visible. Results are written per beam to multibeam_cinr.csv.

Q18. What is in the access report?

Every point analysis produces one, from the same per-satellite time series that drives the link budget:

• Coverage percentage, total accesses, maximum simultaneous satellites
• Access-duration and gap-duration statistics, Time Average Gap, and mean and maximum response time
• A per-satellite visibility Gantt chart and a simultaneous-satellites step chart, with system-level gaps highlighted across both
• Collapsible interval tables with CSV downloads: access_intervals.csv, one row per satellite pass, and gap_intervals.csv, one row per system gap

Q19. What comes out of a run?

On the globe: heatmaps, footprint polygons, satellite markers, and ground links, with playback for synthetic modes at 1x to 50x and timeline scrubbing, and recording controls for live modes.

On disk, CSV throughout: point_analysis.csv, coverage_grid.csv, satellite_footprint.csv, multibeam_cinr.csv, access_intervals.csv, and gap_intervals.csv. The link-quality fields, C/N₀, C/N, G/T, E_b/N₀, BER, and PFD margin, are included in the rows rather than kept behind a separate report. The files are written to be read directly by pandas or Excel.

Q20. How reproducible is a study that used live TLEs?

Live TLEs are saved with the project, together with their capture time and source. Reloading the project with the saved TLEs reproduces the original satellite positions; fetching fresh TLEs instead gives the current positions. A live session can also be recorded and replayed later as captured, with the same playback and scrubbing controls as a synthetic run. Without the stored ephemeris, a result computed against a live constellation cannot be reproduced once the TLEs are updated.

Q21. Can Astra run without the browser, for sweeps and automation?

Yes. The three configuration tabs are saved as one config.json, and the included CLI runs that file headlessly:

cli.py run config.json --set rf.eirp_dbw=60

Any field can be overridden with a dotted path, so a parameter sweep is a shell loop over that command. The CLI returns typed exit codes for use in a CI pipeline. The CLI and the web UI read the same file, so a study configured in the browser runs unchanged on the command line.

Q22. How is Astra deployed for a team?

Astra is web-based and is deployed on your server. It supports multiple users with authentication and per-user projects, and is used from a browser, so there is no desktop install per analyst and no local license tied to a machine. One installation serves the team.

Q23. What is the boundary between NetSim Astra and NetSim NTN?

Astra covers the constellation, the link budget, and the RF coverage, and stops at the radio link. NetSim NTN simulates the 5G NTN protocol stack and the traffic end to end at packet level, reporting throughput, latency, and error performance per beam. Astra outputs define the coverage and link conditions; NTN evaluates protocol and application performance under them.

Q24. How is an evaluation arranged?

Write to sales for an evaluation license. A scenario already run in another tool can be configured in Astra and the outputs compared directly. The NetSim Astra product page covers the feature set in more detail.