VEGA Company Presentation

1
Galileo System Simulation Facility
(GSSF) Validation Approach Frank Zimmermann,
Thomas Haak (VEGA) Chris Hill (University of
Nottingham)Henno Boomkamp, Rene Zandbergen
(ESA/ESOC)
2
Overview

• Introduction
• GSSF Simulation Capabilities
• Modelling Assumptions
• System Level Validation of GSSF
• Independent Validation at ESOC
• Summary

Technical Excellence . Pragmatic Solutions .
Proven Delivery
3
Introduction

• Galileo
• Independent, global, European satellitenavigation
  system
• For civilian applications interoperable with
  GPS
• GSSF
• Simulation environment that reproducesthe
  functional and performance behaviour of
  theGalileo system
• Offers the necessary flexibility and functional
  scopeto support Galileo system simulation needs
  during the entire program life cycle.
• Developed on behalf of ESA/ESTEC by an
  international consortium VEGA (prime), Scisys
  Space and Defence Ltd. (UK), Dataspazio S.p.A.
  (Italy), CAE Inc. (Canada), SENER S.A. (Spain),
  University of Nottingham (UK), IfEN GmbH
  (Germany) and GMV S.A. (Spain)
• Further upgrades currently being carried out by
  VEGA, University of Nottingham, DEIMOS (Spain)
  and TAITUS (Italy)

4
GSSF Simulation Capabilities

• GSSF provides a single simulator that uses
  alternative models depending upon the type of
  analysis the end-user wishes to perform (more
  flexibility than traditional simulators)
• Service Volume Performance Analyses (navigation
  and integrity performance over longer time
  periods and over large geographical areas).
• Raw Data Generation (Galileo and GPS raw data for
  experimental purposes RINEX observation, IGS
  SP3)
• Simulation of the nominal system and its various
  degraded modes (deterministic or probabilistic
  manner)

5
GSSF Model Components

• Space Segmentsatellite models
• Environmentmain perturbationson satellite
  signal
• User Segmentuser receivers andinternal
  algorithms
• Ground SegmentGS modelsfollowing
  Galileobaseline

6
Modelling Assumptions

• SVS
• Model fidelity satisfies required accuracy while
  maintaining sufficient runtime performance.
• Keplerian orbit prediction
• Environmental delays are derived from UERE
  Budgets.
• Optimised for simulation over large geographical
  areas and long time periods
• RDG
• High fidelity environmental modeling and orbit
  prediction for accurate data provision
• Ionosphere (NeQuick), Troposphere (Hopfield),
  Gravity (JGM3 and third-body perturbations),
  Solar Radiation Pressure, Multipath, Noise,

7
GSSF User Interface

• Integrated graphicalelements are used to
• control the simulation
• provide access tosimulation parameters
• process the results
• Developed onWindows (.NET)
• Models platformindependent (C)
• No run-time dependenceon commercial products

8
SVS Simulation Example (1)

• Nominal SISMAfor Galileo on aGSS network
• Representslevel of accuracyin
  monitoringsatellite positions

9
SVS Simulation Example (2)

• Nominal SISMAfor Galileo on aGSS network
• Subject to twostation failuresover North
  America
• SISMA degradesaccordingly

10
GSSF System Level Validation

• SVS Validation
• Against simulated data from trusted sources
• RDG Validation against real data
• RINEX files from Kourou were extracted from the
  GSTB-V1 test data set together with IGS ephemeris
  and clock files.
• Code, carrier and Doppler measurements as
  simulated by GSSF RDG were compared with those
  obtained from the real Kourou data.
• The RDG Validation process adopted an incremental
  approach, where stepwise, external data files
  were replaced by GSSF models.

11
RDG Validation Stepwise Approach

• Step 1 - Baseline scenario Replace the GSSF
  models with data sources which represent the
  prevailing conditions in the measured data from
  the TDS
• ? minimise the differences between the measured
  and simulated data
• Step 2 - Incremental tests Incrementally replace
  these real data sources with the corresponding
  GSSF models. At each step, the increasing
  differences between the real data and the
  simulated data are assessed.
• ? characterisation of the level of fidelity that
  can be expected from GSSF
• Step 3 - Load test Repeat test for full network
  of Galileo sensor stations
• ? check performance 48 hours, 1 second
  time-step, 28 GPS spacecrafts, 23 Ground
  Stations,1 Channel L1C/A, Observables C1, L1,
  D1, S1 (10 h 30 min)

12
Independent Validation by ESOC (1)

• ESA/ESOC have carried out an independent
  validation of GSSF with the following prime
  objectives
• Characterise the GSSF orbits in terms of model
  fidelity and validity of implementation
• Process Raw Data (RINEX and SP3) produced by GSSF
  as it was real measurement data using the
  standard IGS processing software at ESOC

13
Independent Validation by ESOC (2)

• Rationale and Test Case Definition
• (1) System-level test for propagated orbits and
  Earth rotation model
• (2) System-level test for simulated GPS tracking
  data (code phase)

GSSF propagated orbits forming Earth-fixed XYZ
observations
Different combinations of perturbation models
allow validation of individual GSSF force models
ESOC POD software fits internal orbit models to
external XYZ data
Solution characteristics as similar as feasible
to GPS processing at ESOC for routine IGS
products
GSSF simulated GPS tracking data for 25 ground
stations
ESOC POD performs routine IGS-like POD process on
GPS data
14
Independent Validation by ESOC (3)

• ESOC orbit models vs. GSSF models 64 mm RMS fit
  over 24 hrs
• Tracking data fit post-solution tracking
  residuals are 20 cm RMS for code data and 3 cm
  RMS for phase data

15
Independent Validation by ESOC (4)

• Conclusion
• ESOC was able to fit the GSSF orbits with high
  accuracy.
• The GSSF Raw Data is a valid representation of
  realmeasurement data.
• Independent validation will remain integral part
  of project during ongoing upgrade activities.

16
Summary

• GSSF provides a single simulator that uses
  alternative models depending upon the type of
  analysis the end-user wishes to perform.
• GSSF provides Raw Data Generation for GPS and
  Galileo
• Functional TDS is already available and was
  validated with GPS data.
• Calibration with real GSTB V2 data will allow
  production of Galileo TDS with proven
  representative performance
• Software tools used within the Galileo program
  can be rigorously tested also for off-nominal
  conditions.
• SVS Capability meant primarily for definition can
  be applied to teaching and outreach.

17
www.gssf.info www.vega-group.com frank.zimmermann_at_
vega.de
Consulting and Technology Technical Excellence .
Pragmatic Solutions . Proven Delivery
18
www.gssf.info
19
GSSF RDG Upgrades

• The following RDG upgrade activities are
  currently ongoing
• Ionosphere Modelling (higher order terms and
  residual errors)
• Troposphere Modelling (SINEX Import)
• Antenna Phase Centre Offset
• Code Tracking Error Variance
• HW Biases and Receiver Cycle Slips
• Earth Solid Tides
• Solar Radiation Pressure Model (Galileo-specific)
• RDG Model Calibration against real GSTB V2
  satellite data
• Production of Galileo Test Data Set with
  representative performance

20
RDG Validation Example Result

• Penultimate step of the incremental tests - all
  the external data sources, apart from the SP3
  ephemeris, were replaced with their equivalent
  GSSF models
• ? Majority of the measurements agree to better
  than 2 m.
• ? GSSF is capable of reproducing raw data that is
  very similar to real data.

21
GSSF External Interfaces

• GSSF provides internal andexternal Interfaces
  that allow
• Exchange of algorithmsand external
  dataingestion
• ConstellationInitialisation
• EnvironmentalModels
• Ingest data produced byGSSF into other tools
  forfurther analysis

22
GSSF Data File Format

• Text-based file format
• Human readable
• Used for input and output
• Used internally in GSSF but accessible
• Used for all recorded simulation data
• Input/output for analyses and post processing
• Input for visualisation
• Simplifies data exchange with external tools

23
SVS Modelling Assumptions

• Model fidelity satisfies required accuracy while
  maintaining sufficient runtime performance.
• Keplerian orbit prediction
• Environmental delays are derived from UERE
  Budgets.
• Optimised for simulation over large geographical
  areas and long time periods

24
RDG Modelling Assumptions

• 3 major blocks represent real Galileo Segments.
• (User Segment is modelled by means of user
  receiver locations and internal algorithms but
  not relevant for RDG.)
• High fidelity environmental modelling for
  accurate data provision