Navigation System Volume Simulation based on DPSSF

Available online at www.sciencedirect.com

Procedia Engineering

ProcediaProcedia Engineering 00 (2011) Engineering 16000–000 (2011) 245 – 251 www.elsevier.com/locate/procedia

International Workshop on Automobile, Power and Energy Engineering

Navigation System Volume Simulation based on DPSSF Chao Wang*, Xiaoshuang Wang, Qun Li, Weiping Wang National University of Defense Technology Changsha HunanChina

Abstract To resolve the model integration problem in Navigation System Volume Simulation (SVS), A distributed & parallelized simulation framework based on Simulation Model Portability (SMP2) and Service-Oriented Architecture (SOA) is promoted. The SVS and its main criterions are introduced and then a simulation SVS model framework is propounded. The distributed & parallelized SMP2 simulation framework based on SOA (DPSSF) is detailed. At last, a example of global or regional visibility analysis is given to show how the DPSSF is used in SVS.

© 2010 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of Society for Automobile, Power and Energy Engineering Keywords: SVS; SMP2; SOA

1. Introduction Simulation Analysis and Evaluation system of satellite navigation system's service performance on Positioning, Velocity-measuring and Timing plays an important role in the construction of satellite navigation system. Simulation Analysis and Evaluation system of Satellite navigation system is a large scale complex system, there are many characteristic about this kind of system compared to simple systems: 1) The system development involves many organizations which belong to different county and region or different domain and department. 2) The whole system would be divided to many sub-systems, which be distributed to different organizations to develop.

*

* Corresponding author. Tel.: +86-731-84573558. E-mail address: [email protected].

1877-7058 © 2011 Published by Elsevier Ltd. doi:10.1016/j.proeng.2011.08.1079

2462

al. / Procedia Engineering 16 (2011) 245 – 251 C.Chao WangWang et al./etProcedia Engineering 00 (2011) 000–000

3) Every organization have their own methods, techniques, standard and tools to design, analysis, evaluate and implement the sub systems. So, the model integration in construction of these complex systems becomes very complicated. 4) In the conceptual level, these models were developed by different organizations through different modeling paradigm, it’s hard to achieve common concept and understanding among the model developers and integrators. 5) In the implementation level, these models were developed by different organizations through different computer platforms and programming languages, it’s hard to select a foundational technique platform to integrate all the models together. 6) In the communication level, these models may be distributed in different regions of the world. A distributed model integration method could greatly decrease the difficulty and workload of the whole system development. To resolve the aforementioned problems, two issues should be considered: model standard and simulation system architecture. Simulation Model Portability 2 (SMP2)[1][2][3] is propounded by European Space Agency (ESA) in 2004. Its purpose is to meliorate the portability, maintainability and reusability of the simulation models. SMP2 assimilates the advantage of Component-based Design (CBD) and Model Driven Architecture (MDA), accepts the open standard as United Modeling Language (UML) and Extensible Markup Language (XML), finally provides a model development framework, and related tools. In recent years, Service Oriented Architecture (SOA) and Web Service technologies have been well studied to improve the distributed computing. SOA, which integrates the existing service to achieve the required functionalities to build the application, is an ideal distributed software development paradigm for the complex simulation system.[4] In SOA, simulation components are loosely coupled. They can be discovered and composed to form a simulation application.[5] The components can be deployed to heterogeneous platform and communicate via standard protocol, such as XML, SOAP. Distributed & Parallelized SMP2 Simulation Framework (DPSSF) takes the SMP2 as the model development standard and the SOA as the system development paradigm. DPSSF provides a solution for the Simulation Analysis and Evaluation of complex systems. This paper construct the Navigation System Volume Simulation based on the DPSSF. 2. Navigation System Volume Simulation Navigation System Volume Simulation (SVS) provides the ability to analysis the navigation performance and integrity over large region and long time, it can be used to compare the constellation’s designing schemes and analyze the navigation performance (precision, availability, continuity) of the constellation. During the operational phase of navigation system, it can be used to evaluate the effect of satellite failure, assist the constellation’s extending design and analyze the navigation performance of the constellation integrated with other navigation systems. SVS system supports the analysis of Visibility, Coverage, Geometry, Dilution of Precision (DOP), Navigation System Precision (NSP), etc. The receiver to satellite visibility is an important performance measure of navigation System’s performance. The visible satellites decide the receiver’s selection of satellites composition, which greatly affects the precision to Positioning, Velocity-measuring and Timing. Coverage is a rationality measure of constellation. A reasonable constellation design should max the coverage of earth’s surface contemporarily minimize the quantity of satellites. Geometry analysis contains Receiver to Satellite’s Doppler rate/ velocity, elevation, azimuth, etc. These criterions reflect the constellation design’s influence on the performance of navigation system. DOP is an import performance measure of the rationality and navigation capability of designed

247 3

Chao Wang et etal./ al. /Procedia ProcediaEngineering Engineering00 16(2011) (2011)000–000 245 – 251 C. Wang

constellation. The precision of the positional/temporal solution given by the satellite navigation system is finally expressed as the product of DOP and User Equivalent Rang Error (UERE). The lower DOP is, the higher the precision of the positional/temporal solution is. NSP is the straightest representation of the precision of positioning, velocity-measuring and timing. NSP is decided by two factors: precision of the observation value and the intensity of the geometry graphic which is figured out by the satellite distribution. The lower NSP is, the higher the precision of the positional solution is. To support the navigation system volume simulation, a SVS model framework is propounded, which is shown in figure 1. The details of the algorism of above analysis could be found in reference.[6] model

enumeration

interface

structure

StationModel GroundNetworkModel

algorism

container

record model

reference

attribute

output

-Name : String -Description : String -StationCount : int

0..*

-General -Geographical 0..* -Misc

StationsContainer

-strName : String -nStationID: int -strDescription : String

-fLatitude : double -fLongitude : double -fAltitude : double

-nSatelliteSelection : SelectSatEnum -nUEREComp : UEREEnum -fMaskingAngle : double

PosStruct

ECEFStruct

NSPStruct

DOPStruct

IntegrityStruct

-fx : double -fy : double -fz : double

-fLatitude : double -fLongitude : double -fAltitude : double

-fPNSP: double -fTNSP: double -fHNSP: double -fVNSP: double

-fPDOP: double -fTDOP: double -fGDOP: double -fHDOP: double -fVDOP: double

-fHPL_GIC: double -fVPL_GIC: double -fHPL_RAIM: double -fVPL_RAIM: double

+initEP() +computeEP() UserModel

ConstellationModel

-Name : String -Description : String -SatCount : int

GenStruct

IStation

-strName : String -strDescription : String -nSVNO: int

IUser

+getStationID() : int +getStationPos() : ECEFStruct +getMaskingAngle() : double

+getUserPos() : PosStruct +addVisualSatellite( pSatellite : SatelliteStruct ) : void +cleanVisualSatellite() : boolean +getNextVisualSatellite() : SatelliteStruct +reset() : void +getDOP() : DOPStruct +getNSP() : NSPStruct +getIntegrity() : IntegrityStruct +getService() : ServiceStruct +getMaskingAngle() : double

IUserRef

0..* SatelliteContainer IStationRef

0..* SatelliteModel

-General -Advanced -OrbitalElement -Velocity : PosStruct -Acceleration : PosStruct -Position : PosStruct -nType : ConstellationEnum -bHelthy : boolean +initEP() +computeOrbitEP() -computeOrbit() -computeVelocity() -computeAcceleration()

IUserGrid ISatellite

+getPRN() : int +addStationID( StationID: int ) : void +getVelocity() : PosStruct +getAcceleration() : PosStruct +getHealthy() : boolean +getPosition() : PosStruct +getNextStation() : int +reset() : void +cleanStation() : boolean +getType() : ConstellationEnum

OrbitStruct

-fSemiMajorAxis : double -fPerigee : double -fAscension : double -fAnomaly : double -fEccentricity : double -fInclination : double

+getNextGrid() : UserStruct +getGridDOP( nGridID: int ) : DOPStruct +getGridNSP( nGridID: int ) : NSPStruct +getGridIntegrity( nGridID: int ) : IntegrityStruct

EnvironmentModel ISatelliteRef

-fSignalFreq : double

AdvancedSt

-bHealthy : boolean -nPRNNO: int -nSVType : ConstellationEnum -fSignalFreq : double -fSignalPower : double

RecorderModel

FilterBaseStruct

IServiceGridRef

+writeFileHeader() +finish()

-bFliter : boolean -nStartTime : int -nStopTime : int -nSamplingFactor : int

CoverageRecorderModel

-CoverageFilter

-fPercentHMD: double -fPercentVMD: double -fPercentJA : double -fTheshold : double

VisibilityRecorderModel

GeometryRecorderModel

DOPRecorderModel

NspRecorderModel

IntegrityRecorderModel

-VisibilityFilter

-GeometryFilter

+RecorderEP() +initEP()

+RecorderEP() +initEP()

-DOPFilter -DOPAvailabilityFilter

-NSPFilter -NSPAvailabilityFilter -NSPContinuityFilter

-IntegrityFilter -IntegrityAvailabilityFilter -IntegrityContinuityFilter

+RecorderEP() +initEP()

+RecorderEP() +initEP()

+RecorderEP() +initEP()

+RecorderEP() +initEP()

-bFilterStation : boolean -aFliterStationID: Array -bFilterConstellation : boolean -aFilterConstellationType : Array -nVisualStationThreshold : int VisibilityFilter

-nFilterConstellationType : ConstellationEnum -bFilterRegion : boolean -fMaxLatitude : double -fMinLatitude : double -fMaxLongitude : double -fMinLongitude : double

-bFilterRegion : boolean -fMaxLongitude : double -fMinLongitude : double -fMaxLatitude : double -fMinLatitude : double -fA0Percent : double -fA1Percent : double -fA2Percent : double

-ServiceFilter +RecorderEP() +initEP()

DOPFilter GeometryFilter

-bFilterReceiver : boolean -aFilterReceiverList : Array

-bFilterRegion : boolean -fMaxLongitude : double -fMinLongitude : double -fMaxLatitude : double -fMinLatitude : double

InterferenceRecorderModel

-bFilterRegion : boolean -fMaxLongitude : double -fMinLongitude : double -fMaxLatitude : double -fMinLatitude : double

-fGDOPThreshold : double -fPDOPThreshold : double +RecorderEP() -fHDOPThreshold : double +initEP() -fVDOPThreshold : double -bFilterTDOP: boolean -bFilterGDOP: boolean -bFilterPDOP: boolean -bFilterHDOP: boolean NSPFilter -bFilterVDOP: boolean -bFilterRegion : boolean -fMaxLongitude : double -fMinLongitude : double -fMaxLatitude : double -fMinLatitude : double

NSPAvailabilityFilter

-fPNSPThreshold : double -fTNSPThreshold : double -fHNSPThreshold : double -fVNSPThreshold : double -bFilterPNSP: boolean -bFilterTNSP: boolean -bFilterHNSP: boolean -bFilterVNSP: boolean

-UEREValue

computeInterference

-bA0 : boolean -fA1 : double -fA2 : double -nCriticalSatNum: int

-ReceivedPower -ThermalNoise -InterSystem -IntraSystem

-setup -input -output

InterferenceUserStruct

-setup -input -output

-A0 Flag -A1 Percent -A2 percent -CriticalSatNum

-pGeneral : UserBaseStruct -pDOP: DOPStruct -pNSP: NSPStruct -pIntegrity : IntegrityStruct -pInterference : InterferenceStruct

-input -setup -output

-coordinate -VelocityVector -accelerationVector

computeVelocity

OS CS SOL SSRS

computeDopplerShift

<> SelectSatEnum

computeDopplerRate

RxSatSelAllInViewSvs RxSatSelBestDopSvs

RandomSelectionStruct

-bRandomSele ctionAl : boolean -bRandomSelectionA2 : boolean -fSelectionRatioAl : double -fSelectionRatioA2 : double

computeVisual

-VisualFlag -TrueRange -Elevation -Azimuth

-setup -input -output

NSPContinuityFilter

IntegrityContinuityFilter

-nOperationTime : int

-nOperationTime : int

<> UEREEnum TotalUEREmodel ContributionUEREmodel

IntegrityAvailabilityFilter

-fHAL : double -fVAL : double

InterferenceStruct

-fReceivedPower : double -fThermalNoise : double -fInterSystem: double -fIntraSystem: double <> ServiceTypeEnum

computeAcceleration CombinationStruct

-bCombineOnce : boolean -bFailOnlyVisibleSatellites : boolean -nMaxNoOfFailedSatellites : int -nMinNoOfFailedSatellites : int

IntegrityFilter

ServiceRecorderModel -fTDOPThreshold : double

-setup -input -output

computeOrbit AvailabilityStruct

-bCheckNSP: boolean -bCheckPL : boolean -bCheckCriticalSatellites : boolean

+initEP() +computeEP() -computeUERE() -selectSatellites() -computeNSP() -computeIntegrity() -computeService()

DOPAvailabilityFilter

ServiceFilter

CoverageFilter

ServiceUserStruct

-pGeneral : UserBaseStruct -pDOP: DOPStruct -pNSP: NSPStruct -pIntegrity : IntegrityStruct -pService : ServiceStruct

computeUERE

computeService SignalCharStruct

-fSignalToNoise : double -fThermalConstant : double -fRxTemperature : double

-General -Geographical -Misc -ReceiverProperty -Availability -Combination -RandomSelection -Thresholds

IServiceGrid

+getNextGrid() : ServiceUserStruct +getService( nGridID: int ) : ServiceStruct

-HPL GIC -VPL GIC -HPL RAIM -VPL RAIM

ServiceStruct Threshold

ServiceUserGridModel

-General -m_pTimeKeeper -FilterBase

-pGeneral : UserBaseStruct -pDOP: DOPStruct -pNSP: NSPStruct -pIntegrity : IntegrityStruct

computeIntegrity

-fMaskingAngle : double -nReceiverType : RxTypeEnum -fUEREMagin : double -UEREFileName : String -nReceiverServiceType : ServiceTypeEnum -nUserEnvironment : EnvironmentTypeEnum

+initEP() +computeEP() -computeUERE() -selectSatellites() -computeInterference()

-strName : String -strDescription : String -strFileName : String

UserStruct

-PNSP -TNSP -HNSP -VNSP

-setup -input -output

ReceiverPropertyStruct

-General -Geographical -SignalChar -strInterFileName

+getNextGrid() : InterferenceUserStruct +getInterference( nGridID: int ) : InterferenceStruct

UserBaseStruct

-nGridID: int -pGridPos : PosStruct

computeNSP

-nIntegrity : IntegrityEnum -nSatelliteSelection : SelectSatEnum -nUEREComp. : UEREEnum

+initEP() +computeEP() -computeUERE() -selectSatellites() -computeDOP() -computeNSP() -computeIntegrity()

-PDOP -TDOP -GDOP -HDOP -VDOP

-setup -input -output

MiscStruct

InterferenceUserGridModel

IInterferenceGrid

IInterferenceGridRef

computeDOP

-fMaxLatitude : double -fMaxLongitude : double -fMinLatitude : double -fMinLongitude : double -fResolutionX : double -fResolutionY : double

-General -Geographical -Misc -ReceiverProperty -threshold

-SatPRN: int -fSatRxRange : double -fSatRxElevation : double -fSatRxAzimuth : double -fSatDopplerShift : double -fSatDopplerRate : double

-PRNSelected

-setup -input -output

GeographSturct

UserGridModel

IBaseUserGrid

+getGridPos( nGridID: int ) : ECEFStruct +getGridCount() : int +reset() : void +addGridVisualSatellite( nGridID: int, pSatellite : SatelliteStruct ) : void +getNextGridVisualSatellite( nGridID: int ) : SatelliteStruct +resetVisualSatellite( nGridID: int ) : void +cleanVisualSatellite( nGridID: int ) : boolean +hasNextGridVisualSatellite() : boolean +hasNextGrid() : boolean +getMaskingAngle() : double

SatelliteStruct

-setup -input -output

-strName : String -strDescription : String -bCheckDOP: String -bCheckNSP: String -bCheckIntegrity : String

+initEP() +computeEP() -computeUERE() -selectSatellites() -computeIntegrity() -computeDOP() -computeNSP() -computeService()

IUserGridRef

+initEP() -computeDopplerShift() -computeDopplerRate() -computeVisual()

selectStatellites GeneralStruct

-General -Misc -ReceiverProperty -threshold -UserPos : ECEFStruct -DOP: DOPStruct -NSP: NSPStruct -Interference : InterferenceStruct -Service : ServiceStruct -Integrity : IntegrityStruct

ThresholdStruct

-fCriticalSVAlertLimit : double -fHNSPAlertLimit : double -fPNSPAlertLimit : double -fTNSPAlertLimit : double -fVNSPAlertLimit : double -fHPLAlertLimit : double -fVPLAlertLimit : double

<> IntegrityEnum GicOnlySvs RaimOnlySvs GICRAIMSvs

<> ConstellationEnum

<> RxTypeEnum

GPS GLONASS GALILEO BD1 BD2

GPS GLONASS GALILEO BD MIX

<> EnvironmentTypeEnum RuralPedestrian RuralVehicle Fixed Aeronautical

Fig. 1.SVS Model Framework 3. DPSSF: Distributed & Parallelized SMP2 Simulation Framework based on SOA 3.1. System Architecture To enhance the interoperability and composability of the Simulation Model Portability Standard 2 (SMP2) in distributed environment, A Distributed & Parallelized SMP2 Simulation Architecture based on SOA is introduced in [7]. SOA-based SMP2 Simulation is made up of many SMP2 simulation nodes. Every node is made up of SMP2 models, SMP2 simulator, local simulation services and simulation coordinator services. Simulator takes charge of the initialization and destruct of local models. Local simulation services preside over the simulation logging, event schedule and interaction in local simulation node. Simulation control, message

248 4

Chao et Wang et al. / Procedia Engineering 16 (2011) 245 – 251 C. Wang al./ Procedia Engineering 00 (2011) 000–000

exchange and time synchronization among simulation nodes are managed by the simulation coordinator services. The process of SMP2 simulation under SOA is following. 1) User inquires model service information from the service proxy using simulation application client; 2) Model composition and schedule are designed according to the model service information and then transmit to all nodes related to simulation; 3) All models are initialized and the interaction relation between models are established; 4) Simulators run the simulation. 3.2. Hybrid time synchronization method based on interaction graph Simulation based on SOA is more inefficient than simulation base on single computer and local network, because of following reasons: 1) Long distance data transmission over internet brings more time delay; 2) Bandwidth of the internet limits the data transmission speed; 3) XML-based message brings burden of coding and decoding. So, coding, decoding and transmission of simulation messages become the performance bottleneck of the SOA-based SMP2 simulation. One of key solutions to improve the SOA-based SMP2 simulation performance is to minimize the quantity of redundant messages and avoid the simulation rollback. In order to improve the time performance of SMP2 Simulation under Service Oriented Architecture, a hybrid time synchronization method is promoted. Every SMP2 simulation node has simulation coordination service to manage its simulation time advance and message exchange. The hybrid time synchronization method considers the interaction information of every simulation node based on interaction graph, then appoint different time synchronization algorithm to them, to reduce the redundant message and simulation rollback as possible. The main step of the method is: 1) Interaction graph modeling. A graph which describes the interaction relationship between models can be abstracted from SMP2 Assembly and Schedule. 2) Model parallelity evaluation. Every model’s parallelity can be calculated according to the interaction graph. 3) Hybrid time synchronization policy determination. Every simulation node’s time synchronization algorithm can be determined by the node’s model parallelity. More detail about the hybrid time synchronization method based on interaction graph could be found in reference [8]. 4. Example: Global or Regional Visibility Analysis 4.1. Algorism of Global or Regional Visibility Analysis The global and regional visibility analysis outputs the maximum, minimum, and mean number of satellites in view for each node on a latitude and longitude grid over the simulation time period. The user can choose to filter the output returned based on the satellite constellation type. Receives a transmitter and receiver position, their antenna unit vectors and their bore sight cone angles.

249 5

Chao Wang et etal./ al. /Procedia ProcediaEngineering Engineering00 16(2011) (2011)000–000 245 – 251 C. Wang

Converts receiver XYZ to Lat., Long. and Ht. The transmitter to receiver range, elevation and azimuth angle, are calculated from the given positions. The elevation angle is used to determine if the transmitter is visible to the receiver, and by comparison to the azimuth dependent receiver elevation mask, whether it should be used in the position computation, and thus passed through the subsequent objects. The calculation process is shown as follow. 1) True range calculation

TrueRange = dx 2 + dy 2 + dz 2

(1)

Where TrueRange is the geometric distance between two position vectors that relate to the same epoch in the same coordinate system. dx，dy，dz is Difference in x,y,z position between Transmitter and Receiver. Elevation calculation (2) sum = cos(lat ) cos(lon)dx + cos(lat ) sin(lon)dy + sin(lat )dz

 = sin −1 ( sun / TrueRange)

(3) Where lat is latitude of earth bound transmitter or receiver (radians). lon is longitude of earth bound transmitter or receiver (radians). dx, dy, dz is difference in x,y,z position between transmitter and receiver. TrueRange is distance from Transmitter to Receiver (km). Theta is elevation angle (radians). Visibility determinant If Eleveation < ElevationMask Availability flag = NOT_VISIBLE Else Availability flag = VISIBLE Where ElevationMask is decided by the receiver. 4.2. Simulation Configuration The models in SVS model framework were all designed, implemented and deployed in three simulation nodes. As shown in the top left part of the figure 4. For the example of global or regional visibility analysis, SatelliteModel, ConstellationModel, EnvironmentModel, UserGridModel and DataRecordModel were selected. Then five instances of SatelliteModel, one instances of ConstellationModel, one instances of EnvironmentModel, one instances of UserGridModel and one instances of DataRecordModel were created. The parameters of the five satellites were set, and the user grid is set to -50~50 in longitude and -180~180 in latitude, the ElevationMask of the receive was 10°. The detail of model selection, instance creation and composition is shown as the down left of the figure 4. After the model instances composition, four simulation tasks were defined. Simulation duration was 2000-07-12 00:00:00 to 2000-07-12 17:00:00, simulation step was 300 seconds. The detail of model instance schedule is shown as the up right of the figure 4. 4.3. Hybrid time synchronization policy determination The Models, fields, events related to the simulation are given in the table 1. The interaction graph is shown as figure 2. The parallelity values of model instances are given in the table2. The parallelity values of simulation nodes and the time synchronization algorisms are given in the table3. In the tables, MSF is message sending frequency, MRF is message receiving frequency, TSA is time

250 6

Chao et Wang et al. / Procedia Engineering 16 (2011) 245 – 251 C. Wang al./ Procedia Engineering 00 (2011) 000–000

synchronization algorism, SRF is the optimized time synchronized algorithm which forbid the secondary rollback, SAMF is the optimized time synchronized algorithm which forbid the secondary anti-message, TW is the time warp optimized time synchronized algorithm. 4.4. Simulation Result After the model instance composition and schedule, the simulation could run. The simulation result is shown in down right of figure4, in which colours are used to display the average visual satellite quantity of every part of the region. Five time synchronization algorisms are taken to run the same simulation, to analysis the performance of the hybrid time synchronization method. The time of five simulations is shown in figure 3. CMB in the figure is a typical conservative time synchronized algorithm. HTSM is the hybrid time synchronization method. It could be found that CMB algorithm created many redundant messages, so the simulation time performance is the worst. The SRF and SAMF algorithm are better than the TW algorithm. The hybrid method has gained the best time performance, because it fully utilized the advantage of every time synchronization algorithm according to the application. Table 1. Models, fields, events related to the simulation MODELS

FIELDS

EVENT SENT

EVENT RECIEVED

Con Con_time Sat Sat_time Sat_position Env Env_conpositon Env_satposition UG UG_satvisibility UG_conpositon DR DR_satvisibility Table 3. Time synchronization Algorithm of the nodes SIMULATION

MODEL

NODES

DEPLOYMENT

MSF

MRF

TSA

node1

con,sat1,sat2

7

2

SRF

Node2

sat3,sat4,env

3

7

SAMF

Node3

sat5,ug,dr

1

1

TW

Fig.3.Simulation Time

Table 2. Parallelity values of model instances MODEL MSF MRF INSTANCES con

5.0

0

sat1

1

1

sat 2

1

1

sat 3

1

1

sat 4

1

1

sat 5

1

1

env

1

5

ug

1

1

dr

0

1

Fig. 2.Interaction graph

Chao Wang et al. / Procedia Engineering 16 (2011) 245 – 251 C. Wang et al./ Procedia Engineering 00 (2011) 000–000

Fig. 4.Distributed & Parallelized SMP2 Simulation Architecture based on SOA 5. Conclusion Distributed & Parallelized SMP2 Simulation Framework based on SOA takes the SMP2 as the model development standard and the SOA as the system development paradigm, can resolve the integration problem in the Navigation System Volume Simulation. This paper constructs the Navigation System Volume Simulation based on DPSSF, details the model design, selection, composition, schedule, parallelity evaluation and at last shows the simulation result and analyzes the time performance of the five time synchronization algorisms. 6. REFERENCES [1] European Space Agency. SMP 2.0 Handbook Issue 1 Revision 2. EGOS-SIM-GEN-TN-0099, 2005.10. [2] European Space Agency. SMP 2.0 Metamodel Issue 1 Revision 2. EGOS-SIM-GEN-TN-0100, 2005.10. [3] European Space Agency. SMP 2.0 Component Model Issue 1 Revision 2. EGOS-SIM-GEN-TN-0101, 2005.10. [4] THOMAS Erl. Service-Oriented Architecture: Concepts, Technology, and Design. Prentice Hall PTR. 2005.08. [5] Yinong Chen. Modeling and Simulation for and in Service-Oriented Computing Paradigm. SIMULATION, Vol. 83, Issue 1, January 2007 3–6. [6] GSSF Team, "Galileo System Simulation Facility–Algorithms and Models," VEGA Group PLC, Darmstadt, Technical Report, GSSFP2.OM.001, 2004, pp. 19-53. [7] Wang Chao, Song Lili, Li Qun, et al. Research on Distributed & Parallelized SMP2 Simulation Architecture based on Service Oriented Architecture. Journal of System Simulation, 2010, 11. [8] Wang Chao, Li Qun, Wang Weiping. A Time Synchronization Method for SMP2 Simulation under Service Oreinted Architecture. Proceedings of International Conference on Electronic Information and Control Engineering, April, 2011

251 7