Tactical Network Modeling and Simulation Study Swapnaja Ranade
- Slides: 27
Tactical Network Modeling and Simulation Study Swapnaja Ranade Michael Stephano
Agenda n n Scenario description Network model Results and analysis Conclusions
Problem Statement n Observe relationship between physical layer parameters and network layer parameters for a tactical network.
Approach n Model a highly mobile tactical network in OPNET and observe parameters through simulation.
Network Topology and Scale n 2 M 1 A 2 tank companies 28 total nodes n 20 minute simulation duration n 4. 5 km x 4. 0 km playbox n
Company Hierarchy 1 st Platoon 2 nd Platoon 3 rd Platoon HQ
Battle Phase March Phase Alpha Company Bravo Company
Radio Propagation n n n 1 W transmission power Omni-directional antenna Antenna height of 3 meters DTED terrain data used TIREM 4 propagation model Data rate used is 11 Mbps
Network Protocols Used n n 802. 11 b AODV IP Represented by OPNET MANET node model
Mobility Model n M 1 A 2 Characteristics 30 mph mean speed n 17 mph 10% slope n 4. 1 mph 60% slope n n MATLAB Model Incorporates DTED data n Speed linearly interpolated based on slope n MATLAB was used to generate trajectory files n
Simulation Metrics Collected n Physical Layer Received Power n SNR n BER n § Network Layer § Throughput § Delay § Packet Loss Ratio
OPNET Demo
Traffic Model n n Position Updates Sensor Updates Vo. IP Traffic Other Traffic
Packet Tx/Rx Rate
Data Collected Time 0. 0 1. 0 2. 0 3. 0 4. 0 5. 0 6. 0 7. 0 8. 0 9. 0 10. 0 11. 0 12. 0 13. 0 14. 0 15. 0 16. 0 17. 0 18. 0 19. 0 Power 1. 5170 e-09 1. 7490 e-09 2. 0140 e-09 2. 2980 e-09 2. 6500 e-09 3. 1420 e-09 3. 7640 e-09 4. 3330 e-09 4. 8420 e-09 5. 4510 e-09 6. 1950 e-09 8. 1250 e-09 1. 2111 e-08 1. 9789 e-08 3. 8793 e-08 5. 6273 e-08 5. 8653 e-08 3. 6067 e-08 2. 3554 e-08 1. 6607 e-08 SNR 41. 3 38. 8 39. 1 39. 4 39. 6 39. 7 39. 9 40. 0 40. 1 43. 5 43. 8 42. 3 40. 4 40. 7 42. 7 43. 8 41. 0 38. 7 37. 9 BER 0 0 0 0 4. 10 e-11 0 2. 20 e-11 4. 20 e-11 0 1. 19 e-10 6. 75 e-10 0 0 1. 20 e-11 0 Throughput 3. 5738 e+05 7. 3606 e+05 1. 1087 e+06 1. 1307 e+06 1. 1183 e+06 1. 1235 e+06 1. 1168 e+06 1. 1325 e+06 1. 1132 e+06 1. 0724 e+06 1. 0481 e+06 1. 0414 e+06 1. 2502 e+06 1. 1259 e+06 1. 1313 e+06 1. 1161 e+06 1. 1164 e+06 1. 1186 e+06 1. 1191 e+06 1. 1302 e+06 Latency 6. 8342 e-03 4. 5720 e-03 1. 1524 e-02 5. 5088 e-03 5. 6622 e-03 5. 6514 e-03 5. 6525 e-03 5. 4484 e-03 5. 6375 e-03 9. 8462 e-03 3. 1363 e-02 4. 6925 e-02 1. 3544 e-01 5. 7345 e-03 5. 9660 e-03 6. 1210 e-03 5. 9077 e-03 5. 9030 e-03 6. 4405 e-03 6. 0194 e-03 PLR 3. 0769 e-02 1. 9084 e-02 1. 4566 e-02 8. 0000 e-03 5. 3814 e-03 4. 5524 e-03 3. 3827 e-03 2. 5316 e-03 2. 3791 e-03 6. 9111 e-03 1. 3079 e-02 2. 0159 e-02 9. 9167 e-03 9. 0520 e-03 8. 4907 e-03 8. 0972 e-03 7. 6680 e-03 7. 5138 e-03 6. 8883 e-03 6. 5212 e-03
Covariance Matrix x y z a b c σx 2 Cxy Cxz Cxa Cxb Cxc σy 2 Cyz Cya Cyb Cyc σz 2 Cza Czb Czc σa 2 Cab Cac σb 2 Cbc σc 2
Correlation Matrix x y z a b c 1. 0 rxy rxz rxa rxb rxc 1. 0 ryz rya ryb ryc 1. 0 rza rzb rzc 1. 0 rab rac 1. 0 rbc 1. 0
Correlation Expectations Prx SNR BER Tput Latency PLR 1. 0 + - - 1. 0 - + + 1. 0 - - 1. 0 + 1. 0
Correlation Matrix Based on OPNET Sim Data Prx SNR BER Tput Latency PLR 1. 0 0. 5512 -0. 0868 0. 2597 -0. 1292 -0. 3624 1. 0 -0. 4596 0. 3374 -0. 1906 -0. 3843 1. 0 0. 0566 -0. 0359 0. 1035 1. 0 -0. 4851 -0. 8758 1. 0 -0. 4573 1. 0
Linear Regression (Native Units) Power vs. Packet Loss Ratio y = mx + b m = -2. 48 x 106 b = 0. 0935
Linear Regression (Z Scores) Power vs. Packet Loss Ratio y = mx + b m = -0. 3624 b=0
Linear Regression SNR vs. Throughput y = mx + b m = 9. 62 x 103 b = 5. 41 x 105
Linear Regression Throughput vs. Packet Loss Ratio y = mx + b m = 9. 62 x 103 b = 5. 41 x 105
OPNET Model Issues Discovered By Analysis n SNR model Mathematical issue n Correctable in OPNET with modifications n n BER model Inadequate OPNET modulation curves n Correctable in OPNET with modifications n
Modulation Curve Produced Analytically r = -0. 5621
Conclusions n Correlation analysis Power and SNR exhibit fairly strong correlation to network parameters n Low BER correlation due to model, not real effect n n Regression analysis n n Useful but of limited utility Choosing appropriate physical layer data
References n n n G. Kelsch, M. Roberts, and D. Harris. Toward New Tactical Network Modeling and Simulation Approaches. In IEEE Military Communications Conference, pages 2396 -2401 Vol. 4, October 2005. United States Army Field Manual. FM 17 -15 Tank Platoon. April 1996. M. J. Ryan and M. R. Frater. Tactical Communications for the Digitized Battlefield. Artech House, Norwood, MA, 2002. S. K. Kachigan. Multivariate Statistical Analysis: A Conceptual Introduction, 2 nd Edition. Radius Press, New York, NY, 1991. General Dynamics M 1 A 2 Abrams Specifications Sheet.
- Gireeja ranade anant sahai
- Abhiram ranade
- Helen c. erickson
- Simulation kelton
- Pharmaceutical simulation and modeling
- Introduction to modeling and simulation
- Biomedical modeling and simulation
- Pengertian pemodelan dan simulasi
- Modeling and simulation
- Relational vs dimensional data modeling
- Nature of simulation
- Basic simulation modeling
- S o f t w a r e f o r t r a f f i c
- Network modeling tools
- Tactical information systems
- Activity resource usage model and tactical decision making
- Activity resource usage model and tactical decision making
- Alexander osterwalder
- Virtual circuit and datagram networks
- Features of peer to peer network and client server network
- Network centric computing and network centric content
- 5 steps in teaching tactical skills
- Tactical information system
- Tactical decision making
- Tactical communication skills
- Integrating communications assessment and tactics
- Nagios tactical overview
- Sec 555