4. TITLE AND SUBTITLE 5. FUNDING NUMBERS A COMPARISON OF ANA

NAVAL POSTGRADUATE SCHOOL Monterey, California

THESIS

A COMPARISON OF ANALYSIS IN DIS AND HLA

by

Steven D. Knight

June 1998

Thesis Advisor: Arnold H. Buss

Second Reader:William S. Murphy, Jr.

Approved for public release; distribution is unlimited.

REPORT DOCUMENTATION PAGE Form Approved

OMB No. 0704-0188 Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources,

gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this

collection of information, including suggestions for reducing this burden to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson

Davis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188), Washington, DC 20503.

1. AGENCY USE ONLY (Leave Blank)

2. REPORT DATE

June 19983. REPORT TYPE AND DATES COVERED

Master’s Thesis

4. TITLE AND SUBTITLE

A COMPARISON OF ANALYSIS IN DIS AND HLA

5. FUNDING NUMBERS

6. AUTHOR(S)

Steven D. Knight

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) Naval Postgraduate School

Monterey, CA 93943-50008. PERFORMING ORGANIZATION REPORT NUMBER

9. SPONSORING / MONITORING AGENCY NAME(S) AND ADDRESS(ES) TRADOC Analysis Center –Monterey

Monterey, CA 93943-510110. SPONSORING / MONITORING AGENCY REPORT NUMBER

11. SUPPLEMENTARY NOTES

The views expressed in this thesis are those of the author and do not reflect the official policy or position of the Department of Defense or the U.S. Government.

12a. DISTRIBUTION / AVAILABILITY STATEMENT

Approved for public release, distribution is unlimited

12b. DISTRIBUTION CODE

13. ABSTRACT (Maximum 200 words)

As the Department of Defense (DoD) continually relies more on Modeling and Simulation (M & S) for testing, analyzing, and training, issues of interoperability have become one of the most important concerns. As such, DoD adopted the Distributed Interactive Simulation (DIS) protocol in 1991. Although successful in many aspects, DIS is limited by available information from models, memory and network requirements, and analytical tools available. Therefore, in 1996 the Defense Modeling and Simulation Office (DMSO) released the High Level Architecture (HLA), an object-oriented approach to interoperability.

This thesis compares these different approaches to analysis to determine functionality in terms of gathering, processing, and reporting on analytical questions in both environments. To compare DIS and HLA analysis, three simulation runs were conducted: Janus vs. Janus in DIS, HLA without an Analysis Federate, and HLA with an Analysis Federate. The Analysis Federate is an HLA-compliant software package that gathers and processes information for analysis requirements. The results of the three simulation runs and subsequent analysis demonstrated the techniques and approaches for each infrastructure. The resulting comparison between them show HLA with the Analysis Federate is the easiest and most functional tool.

The Analysis Federate fills an analysis void currently in HLA and by implementing it with the study question model tree methodology, an analyst will be more effective and be able to provide real-time feedback.

14. SUBJECT TERMS

Analysis Federate, Distributed Interactive Simulation (DIS), High Level Architecture (HLA), Federation Object Model (FOM), Simulation Object Model (SOM), Janus, Gateway, Study Question Model Tree, Federation,15. NUMBER OF PAGES

97

Run-Time Infrastructure (RTI), Protocol Data Unit (PDU), PDU Adapter Software System16. PRICE CODE

i

(PASS)

17. SECURITY CLASSIFICATION

OF REPORT

Unclassified 18. SECURITY

CLASSIFICATION

OF THIS PAGE

Unclassified

19. SECURITY

CLASSIFICATION

OF ABSTRACT

Unclassified

20. LIMITATION OF

ABSTRACT

UL

NSN 7540-01-280-5500 Standard Form 298 (Rev. 2-89)

Prescribed by ANSI Std. 239-18

298-102

ii

ii

Approved for public release; distribution is unlimited.

A COMPARISON OF ANALYSIS IN DIS AND HLA

Steven D. Knight

Captain, United States Army

B.S., United States Military Academy, 1988

Submitted in partial fulfillment

of the requirements for the degree of

MASTER OF SCIENCE IN OPERATIONS RESEARCH

from the

NAVAL POSTGRADUATE SCHOOL

June 1998

Author: _________________________________________________________

Steven D. Knight

Approved by: _________________________________________________________

Arnold H. Buss, Thesis Advisor

__________________________________________________________

William S. Murphy, Second Reader

___________________________________________________________

Richard E. Rosenthal, Chairman

Department of Operations Research

iii

iv

ABSTRACT

As the Department of Defense (DoD) continually relies more on Modeling and Simulation (M & S) for testing, analyzing, and training, issues of interoperability have become one of the most important concerns. As such, DoD adopted the Distributed Interactive Simulation (DIS) protocol in 1991. Although successful in many aspects, DIS is limited by available information from models, memory and network requirements, and analytical tools available. Therefore, in 1996 the Defense Modeling and Simulation Office (DMSO) released the High Level Architecture (HLA), an object-oriented approach to interoperability.

This thesis compares these different approaches to analysis to determine functionality in terms of gathering, processing, and reporting on analytical questions in both environments. To compare DIS and HLA analysis, three simulation runs were conducted: Janus vs. Janus in DIS, HLA without an Analysis Federate, and HLA with an Analysis Federate. The Analysis Federate is an HLA-compliant software package that gathers and processes information for analysis requirements. The results of the three simulation runs and subsequent analysis demonstrated the techniques and approaches for each infrastructure. The resulting comparison between them show HLA with the Analysis Federate is the easiest and most functional tool.

The Analysis Federate fills an analysis void currently in HLA and by implementing it with the study question model tree methodology, an analyst will be more effective and be able to provide real-time feedback.

v

vi

THESIS DISCLAIMER

The reader is cautioned that computer programs developed in this research may not have been exercised for all cases of interest. While every effort has been made, within the time available, to ensure that the programs are free of computational and logic errors, they cannot be considered validated. Any application of these programs without additional verification is at the risk of the user.

vii

viii

TABLE OF CONTENTS

I. INTRODUCTION (1)

A. BACKGROUND (1)

1. General (1)

a. Modeling and Simulation (1)

b. History (Stand-alone to Distributed Simulations) (3)

2. Distributed Interactive Simulation (DIS) (5)

3. High Level Architecture (HLA) (6)

B. JANUS COMBAT SIMULATION MODEL (8)

1. Stand-alone (9)

2. PDU Adapter Software System (PASS) (10)

C. PROBLEM DESCRIPTION (12)

1. Analysis Tools (12)

a. Post Processors (12)

b. Data Loggers (DIS) (13)

c. Analysis Federate (HLA) (14)

2. Advantages and Limitations of the Analysis Tools Available (15)

3. Analytical Approaches in DIS and HLA (17)

D. THESIS STATEMENT AND OUTLINE (17)

II. METHODOLOGY (19)

A. COMPARISON OF ANALYTICAL APPROACHES (19)

1. Janus vs. Janus in DIS (19)

2. Janus vs. Janus in HLA without an Analysis Federate (21)

3. Janus vs. Janus in HLA with an Analysis Federate (21)

B. SCENARIO DEVELOPMENT (22)

ix

C. ANALYSIS QUESTION DEVELOPMENT (23)

D. SOFTWARE AND HARDWARE REQUIREMENTS (25)

1. Janus as the Base Model (25)

2. DIS Setup (26)

3. HLA Setup (27)

a. HLA Gateway (30)

b. Analysis Federate (30)

III. RESULTS (33)

A. SCENARIO DEVELOPMENT (33)

1. Exercise Scenario (33)

a. General Situation (34)

b. Specific Situation (34)

c. Current Situation (35)

2. General (36)

B. ANALYSIS QUESTIONS (40)

C. COMPARATIVE APPROACHES (46)

1. Janus vs. Janus in DIS (46)

a. Data Collection (46)

b. Results (49)

2. Janus vs. Janus in HLA without an Analysis Federate (54)

a. Data Collection (55)

b. Results (56)

3. Janus vs. Janus in HLA with an Analysis Federate (58)

a. Data Collection (59)

b. Results (63)

D. DISCUSSION (65)

IV. CONCLUSIONS AND RECOMMENDATIONS (70)

A. CONCLUSIONS (70)

x

B. RECOMMENDATIONS (72)

C. FUTURE WORK (72)

APPENDIX A. PROTOCOL DATA UNIT (PDU) FORMAT (74)

APPENDIX B. JANUS DATABASE (78)

APPENDIX C. JANUS POST PROCESSOR REPORTS (80)

APPENDIX D. HLA GATEWAY MODIFICATIONS (84)

APPENDIX E. SCENARIO SIDE ONE AND TWO BREAKDOWN (86)

APPENDIX F. PERL CODE FOR KILLER/VICTIM DATA (88)

APPENDIX G. PERL CODE FOR ROUND TYPE DATA (92)

APPENDIX H. GLOSSARY OF ACRONYMS (94)

LIST OF REFERENCES (96)

INITIAL DISTRIBUTION LIST (98)

xi

xii

LIST OF FIGURES

F IGURE 1: PASS IN A DIS ENVIRONMENT [9] (11)

F IGURE 2: S TUDY Q UESTION M ODEL T REE (25)

F IGURE 3: DIS S ETUP (27)

F IGURE 4: HLA S ETUP (29)

F IGURE 5: I NITIAL U.S. F ORCE D ISPOSITION (38)

F IGURE 6: I NITIAL S ERBIAN F ORCE D ISPOSITION (39)

F IGURE 7: S TUDY Q UESTIONS (41)

F IGURE 8: U NIT S TRENGTH MOE S AND D ATA C OLLECTION R EQUIREMENTS (42)

F IGURE 9: U NIT’S LER (43)

F IGURE 10: C LASS V (A MMUNITION) S TATUS (44)

F IGURE 11: C LASS III (F UEL) S TATUS (45)

F IGURE 12: J ANUS VS. J ANUS IN HLA WITHOUT AN A NALYSIS F EDERATE (54)

F IGURE 13: J ANUS VS. J ANUS IN HLA WITH AN A NALYSIS F EDERATE (59)

F IGURE 14: O PERATIONAL V IEW WITH F OCUS S ETS (61)

F IGURE 15: P RIMARY L IST FOR A F OCUS S ET (62)

F IGURE 16: J ANUS D ATABASE H IERARCHICAL D IAGRAM. [15] (78)

xiii

xiv

LIST OF TABLES

T ABLE 1: K ILLER/V ICTIM S COREBOARD (49)

T ABLE 2: R OUND C OUNT BY E NTITY (50)

T ABLE 3: P ERCENT R EMAINING FOR EACH P LATOON IN A C OMPANY (51)

T ABLE 4: A C OMPANY’S LER (52)

T ABLE 5: A MMUNITION E XPENDITURE (53)

T ABLE 6: A C OMPANY’S S TRENGTH (56)

T ABLE 7: A C OMPANY’S LER (56)

T ABLE 8: A C OMPANY’S A MMUNITION E XPENDITURE (58)

T ABLE 9: A C OMPANY’S S TRENGTH R EMAINING (64)

T ABLE 10: A C OMPANY’S LER (64)

T ABLE 11: A C OMPANY’S A MMUNITION E XPENDITURE (65)

T ABLE 12: S TATISTICS FOR O VERALL D ATABASE. [15] (79)

xv

xvi

EXECUTIVE SUMMARY

As the Department of Defense (DoD) continually relies more on Modeling and Simulation (M & S) for testing, analysis, and training, issues of interoperability have become one of the most important concerns. Beginning with the success of the Simulation Networking (SIMNET) program in 1984, in which models interacted in a distributed environment, DoD has continually incorporated emerging technologies to improve interoperability in a distributed environment.

The first DoD-wide standard was the Distributed Interactive Simulation (DIS) protocol adopted in 1991. Under DIS, models broadcast Protocol Data Units (PDUs) over an area network and received PDUs from other models. The PDUs attempted to contain sufficient information to allow various models the ability to represent entities and events within each model. Although successful in many aspects, DIS is limited by available information from models, memory and network requirements, and analytical tools available. Therefore, in 1996 the Defense Modeling and Simulation Office (DMSO) released the High Level Architecture (HLA), an object-oriented approach to interoperability. HLA requires models, or federates, to publish and subscribe to objects, interactions, attributes, and parameters specified in the Federation Object Model (FOM).

Analysis is conducted differently in DIS and HLA due to the differences between broadcasting and publish/subscription requirements. This thesis compares these different approaches to analysis to determine functionality in terms of gathering, processing, and reporting on analytical questions in both environments.

The PDUs broadcast in DIS are sent over a User Datagram Protocol/Internet Protocol (UDP/IP) network. This means any model listening to the right port can receive

xvii

all of the PDUs, and hence exercise information. By recording the PDUs with the use of a data logger, information can be stored and data collection requirements extracted after the exercise is complete. Therefore, analysis questions are typically not developed or data requirements determined until after the exercises.

HLA uses a different approach than DIS. In HLA, a model registering with the federation publishes and subscribes to required information. No analytical tools are inherent in HLA, currently leaving only the inpidual models post processor reports. However, an Analysis Federate has been proposed to perform analysis under HLA. This federate would subscribe to information required to answer analysis questions and measures of effectiveness (MOEs). Additionally, this required information must be known prior to registering, meaning that the analysis questions must also be developed prior to registering.

A proposed question development process is called the study question model tree and was used in this thesis. The study question model tree begins with the overall objective of the exercise and works down through study questions and MOEs until the required data is determined. Then the subscription requirements can be determined from the required data. Once the data is gathered during the exercise, the process reverses until all the study questions have been answered and the objective met.

To compare DIS and HLA analysis, three simulation runs were conducted: Janus vs. Janus in DIS, HLA without an Analysis Federate, and HLA with an Analysis Federate. Each simulation run used the same scenario and analysis questions. The scenario was based on Bosnia and incorporated the factors of realism, flexibility, and

xviii

tactical soundness. The analysis requirements were developed using the study question model tree methodology and used for all three simulation runs.

The results of the three simulation runs and subsequent analysis demonstrated the techniques and approaches for each infrastructure. The resulting comparison between them show HLA with the Analysis Federate is the easiest and most functional tool. It provides a workstation that an analyst can learn to use in a short amount of time and still present quality results. It also provides the opportunity for real-time analysis. This is a big advantage over the other techniques since feedback can be provided to the commanders while the exercise is still executing.

The overall recommendations from this study are twofold. First of all, incorporate the Analysis Federate into all HLA federation requiring analysis. The Analysis Federate developed by the TRADOC Analysis Center (TRAC) – Monterey provides the added functionality of interoperability within any federation. The second recommendation is incorporate the study question model tree methodology to approaching analysis, resulting in a more proactive analyst.

As DoD continues to progress towards HLA, further study on time latency issues, data processing in the Run-Time Infrastructure (RTI), and standardized reports in the Analysis Federate deserve consideration. Each of these areas impact on the overall results of the simulation run by either increasing the accuracy or reducing the amount of processing and calculations that would otherwise be necessary external to the Analysis Federate.

xix

本文来源：https://www.bwwdw.com/article/9bwl.html