This book describes the benefits of sensor fusion as illustrated by considering the characteristics of infrared, microwave, and millimeter-wave sensors, including the influence of the atmosphere on their performance, sensor system application scenarios that may limit sensor size but still require high resolution data, and the attributes of data fusion architectures and algorithms. The data fusion algorithms discussed in detail include classical inference, Bayesian inference, Dempster-Shafer evidential theory, artificial neural networks, voting logic as derived from Boolean algebra expressions, fuzzy logic, and detection and tracking of objects using only passively acquired data. A summary is presented of the information required to implement each of the data fusion algorithms discussed. Weather forecasting, Earth resource surveys that use remote sensing, vehicular traffic management, target classification and tracking, military and homeland defense, and battlefield assessment are some of the applications that will benefit from the discussions of signature-generation phenomena, sensor fusion architectures, and data fusion algorithms provided in this text.
**Contents**
- List of Figures
- List of Tables
- Preface
- Introduction
- Multiple Sensor System Applications, Benefits, and Design Considerations
- Data Fusion Algorithms and Architectures
- Classical Inference
- Bayesian Inference
- Dempster-Shafer Evidential Theory
- Artificial Neural Networks
- Voting Logic Fusion
- Fuzzy Logic and Fuzzy Neural Networks
- Passive Data Association Techniques for Unambiguous Location of Targets
- Retrospective Comments
- Appendix A: Planck Radiation Law and Radiative Transfer
- Appendix B: Voting Fusion with Nested Confidence Levels
- Index

lgli/A:\knovel\3511.pdf

lgrsnf/A:\knovel\3511.pdf

nexusstc/Sensor and Data Fusion: A Tool for Information Assessment and Decision Making/8e1a57dec93a554b5429b904a49f15e8.pdf

PDFsharp 1.32.2608-g (www.pdfsharp.net)

by Lawrence A. Klein

4<8=8AB@0B>@

SPIE--The International Society for Optical Engineering; SPIE Press

Spie Publications

Elsevier

Society of Photo-Optical Instrumentation Engineers (SPIE), [N.p.], 2004

SPIE Press monograph, PM138, Bellingham, Wash, ©2004

United States, United States of America

Bellingham, WA, United States, 2004

July 2004

3511

lg1007109

producers:
PDFsharp 1.32.2608-g (www.pdfsharp.net)

{"isbns":["0819454354","1615837248","9780819454355","9781615837243"],"last_page":342,"publisher":"SPIE","series":"SPIE Press Monograph Vol. PM138"}

Front Matter 1
List of Figures 3
List of Tables 7
Preface 11
Table of Contents 13
1. Introduction 19
2. Multiple Sensor System Applications, Benefits, and Design Considerations 25
2.1 Data Fusion Applications to Multiple Sensor Systems 26
2.2 Selection of Sensors 28
2.3 Benefits of Multiple Sensor Systems 34
2.4 Influence of Wavelength on Atmospheric Attenuation 36
2.5 Fog Characterization 39
2.6 Effects of Operating Frequency on MMW Sensor Performance 40
2.7 Absorption of MMW Energy in Rain and Fog 41
2.8 Backscatter of MMW Energy from Rain 43
2.9 Effects of Operating Wavelength on IR Sensor Performance 46
2.10 Visibility Metrics 47
2.10.1 Visibility 47
2.10.2 Meteorological Range 48
2.11 Attenuation of IR Energy by Rain 49
2.12 Extinction Coefficient Values 50
2.13 Summary of Attributes of Electromagnetic Sensors 50
2.14 Atmospheric and Sensor System Computer Simulation Models 56
2.14.1 LOWTRAN Attenuation Model 56
2.14.2 FASCODE and MODTRAN Attenuation Models 58
2.14.3 EOSAEL Sensor Performance Model 59
2.15 Summary 61
References 64
3. Data Fusion Algorithms and Architectures 68
3.1 Definition of Data Fusion 68
3.2 Level 1 Processing 71
3.2.1 Detection, Classification, and Identification Algorithms for Data Fusion 72
3.2.1.1 Parametric Techniques 75
3.2.1.2 Information Theoretic Techniques 80
3.2.1.3 Logical Templates 85
3.2.1.4 Knowledge-Based Expert Systems 86
3.2.1.5 Fuzzy Set Theory 87
3.2.2 State Estimation and Tracking Algorithms for Data Fusion 88
3.2.1.1 Data Alignment 89
3.2.1.2 Data and Object Correlation 89
3.2.1.3 Prediction Gates 89
3.2.1.4 Association Metrics 89
3.2.1.5 Data Association 89
3.2.1.6 Track-to-Track Association 92
3.2.1.7 Positional, Kinematic, and Attribute Estimation 93
3.2.1.8 Track Initiation 94
3.3 Level 2, 3, and 4 Processing 95
3.4 Data Fusion Processor Functions 97
3.5 Definition of an Architecture 99
3.6 Data Fusion Architectures 100
3.6.1 Sensor-Level Fusion 100
3.6.2 Central-Level Fusion 104
3.6.3 Hybrid Fusion 106
3.6.4 Pixel-Level Fusion 108
3.6.5 Feature-Level Fusion 109
3.6.6 Decision-Level Fusion 110
3.7 Sensor Footprint Registration and Size Considerations 110
3.8 Summary 112
References 114
4. Classical Inference 118
4.1 Estimating the Statistics of a Population 119
4.2 Interpreting the Confidence Interval 120
4.3 Confidence Interval for a Population Mean 122
4.4 Significance Tests for Hypotheses 126
4.5 The z-Test for a Population Mean 126
4.6 Tests with Fixed Significance Level 129
4.7 The t-Test for a Population Mean 131
4.8 Caution in Use of Significance Tests 134
4.9 Inference as a Decision 135
4.10 Summary 139
References 142
5. Bayesian Inference 143
5.1 Bayes' Rule 143
5.2 Bayes' Rule in Terms of Odds Probability and Likelihood Ratio 146
5.3 Direct Application of Bayes' Rule to Cancer Screening Test Example 148
5.4 Comparison of Bayesian Inference with Classical Inference 149
5.5 Application of Bayesian Inference to Fusing Information from Multiple Sources 151
5.6 Combining Multiple Sensor Information Using the Odds Probability Form of Bayes' Rule 152
5.7 Recursive Bayesian Updating 153
5.8 Posterior Calculation Using Multivalued Hypotheses and Recursive Updating 155
5.9 Enhancing Underground Mine Detection Using Two Sensors That Produce Data Generated by Uncorrelated Phenomena 158
5.10 Summary 162
References 163
6. Dempster-Shafer Evidential Theory 164
6.1 Overview of the Process 164
6.2 Implementation of the Method 165
6.3 Support, Plausibility, and Uncertainty Interval 166
6.4 Dempster's Rule for Combination of Multiple Sensor Data 171
6.4.1 Dempster's Rule with Empty Set Elements 173
6.4.2 Dempster's Rule When Only Singleton Propositions are Reported 174
6.5 Comparison of Dempster-Shafer with Bayesian Decision Theory 176
6.5.1 Dempster-Shafer-Bayesian Equivalence Example 177
6.5.2 Dempster-Shafer-Bayesian Computation Time Comparisons 178
6.6 Probabilistic Models for Transformation of Dempster-Shafer Belief Functions for Decision Making 178
6.6.1 Pignistic Transferable Belief Model 179
6.6.2 Plausibility Transformation Function 181
6.6.3 Modified Dempster-Shafer Rule of Combination 186
6.7 Summary 193
References 195
7. Artificial Neural Networks 197
7.1 Applications of Artificial Neural Networks 198
7.2 Adaptive Linear Combiner 198
7.3 Linear Classifiers 199
7.4 Capacity of Linear Classifiers 201
7.5 Nonlinear Classifiers 202
7.5.1 Madaline 202
7.5.2 Feedforward Network 204
7.6 Capacity of Nonlinear Classifiers 206
7.7 Supervised and Unsupervised Learning 207
7.8 Supervised Learning Rules 209
7.8.1 mu-LMS Steepest Descent Algorithm 210
7.8.2 alpha-LMS Error Correction Algorithm 210
7.8.3 Comparison of the mu-LMS and alpha-LMS Algorithms 211
7.8.4 Madaline I and II Error Correction Rules 211
7.8.5 Perceptron Rule 212
7.8.6 Backpropagation Algorithm 214
7.8.7 Madaline III Steepest Descent Rule 216
7.8.8 Dead Zone Algorithms 217
7.9 Generalization 218
7.10 Other Artificial Neural Networks and Processing Techniques 219
7.11 Summary 221
References 227
8. Voting Logic Fusion 229
8.1 Sensor Target Reports 231
8.2 Sensor Detection Space 232
8.2.1 Venn Diagram Representation of Detection Space 232
8.2.2 Confidence Levels 232
8.2.3 Detection Modes 233
8.3 System Detection Probability 235
8.3.1 Derivation of System Detection and False Alarm Probability for Nonnested Confidence Levels 235
8.3.2 Relation of Confidence Levels to Detection and False Alarm Probabilities 237
8.3.3 Evaluation of Conditional Probability 238
8.3.4 Establishing False Alarm Probability 239
8.3.5 Calculating System Detection Probability 240
8.3.6 Summary of Detection Probability Computation Model 240
8.4 Application Example without Singleton Sensor Detection Modes 241
8.4.1 Satisfying the False Alarm Probability Requirement 242
8.4.2 Satisfying the Detection Probability Requirement 242
8.4.3 Observations 244
8.5 Hardware Implementation of Voting Logic Sensor Fusion 245
8.6 Application Example with Singleton Sensor Detection Modes 245
8.7 Comparison of Voting Logic Fusion with Dempster-Shafer Evidential Theory 247
8.8 Summary 248
References 250
9. Fuzzy Logic and Fuzzy Neural Networks 251
9.1 Conditions under which Fuzzy Logic Provides an Appropriate Solution 251
9.2 Illustration of Fuzzy Logic in an Automobile Antilock Braking System 252
9.3 Basic Elements of a Fuzzy System 253
9.4 Fuzzy Logic Processing 253
9.5 Fuzzy Centroid Calculation 255
9.6 Balancing an Inverted Pendulum with Fuzzy Logic Control 257
9.6.1 Conventional Mathematical Solution 257
9.6.2 Fuzzy Logic Solution 259
9.7 Fuzzy Logic Applied to Multitarget Tracking 262
9.7.1 Conventional Kalman Filter Approach 263
9.7.2 Fuzzy Kalman Filter Approach 265
9.8 Fuzzy Neural Networks 270
9.9 Fusion of Fuzzy-Valued Information from Multiple Sources 272
9.10 Summary 273
References 275
10. Passive Data Association Techniques for Unambiguous Location of Targets 277
10.1 Data Fusion Options 277
10.2 Received-Signal Fusion 279
10.2.1 Coherent Processing Technique 281
10.2.2 System Design Issues 283
10.3 Angle Data Fusion 285
10.3.1 Solution Space for Emitter Locations 286
10.3.2 Zero-One Integer Programming Algorithm Development 289
10.3.3 Relaxation Algorithm Development 294
10.4 Decentralized Fusion Architecture 296
10.4.1 Local Optimization of Direction Angle Track Association 297
10.4.2 Global Optimization of Direction Angle Track Association 299
10.4.2.1 Closest Approach Distance Metric 299
10.4.2.2 Hinge Angle Metric 300
10.5 Passive Computation of Range Using Tracks from a Single Sensor Site 301
10.6 Summary 302
References 304
11. Retrospective Comments 306
Appendices 311
Appendix A: Planck Radiation Law and Radiative Transfer 311
A.1 Planck Radiation Law 311
A.2 Radiative Transfer Theory 313
References 317
Index 318
A 318
B 320
C 321
D 323
E 325
F 326
G 328
H 329
I 329
J 330
K 331
L 331
M 332
N 334
O 335
P 336
Q 337
R 338
S 339
T 342
U 343
V 344
W 344
Z 345
About the Author 346

This Book Illustrates The Benefits Of Sensor Fusion By Considering The Characteristics Of Infrared, Microwave, And Millimeter-wave Sensors, Including The Influence Of The Atmosphere On Their Performance. Applications That Benefit From This Technology Include: Vehicular Traffic Management, Remote Sensing, Target Classification And Tracking- Weather Forecasting- Military And Homeland Defense. Covering Data Fusion Algorithms In Detail, Klein Includes A Summary Of The Information Required To Implement Each Of The Algorithms Discussed, And Outlines System Application Scenarios That May Limit Sensor Size But That Require High Resolution Data. Chapter 1. Introduction -- Chapter 2. Multiple Sensor System Applications, Benefits, And Design Considerations -- 2.1. Data Fusion Applications To Multiple Sensor Systems -- 2.2. Selection Of Sensors -- 2.3. Benefits Of Multiple Sensor Systems -- 2.4. Influence Of Wavelength On Atmospheric Attenuation -- 2.5. Fog Characterization -- 2.6. Effects Of Operating Frequency On Mmw Sensor Performance -- 2.7. Absorption Of Mmw Energy In Rain And Fog -- 2.8. Backscatter Of Mmw Energy From Rain -- 2.9. Effects Of Operating Wavelength On Ir Sensor Performance -- 2.10. Visibility Metrics -- 2.10.1. Visibility -- 2.10.2. Meteorological Range -- 2.11. Attenuation Of Ir Energy By Rain -- 2.12. Extinction Coefficient Values (typical) -- 2.13. Summary Of Attributes Of Electromagnetic Sensors -- 2.14. Atmospheric And Sensor System Computer Simulation Models -- 2.14.1. Lowtran Attenuation Model -- 2.14.2. Fascode And Modtran Attenuation Models -- 2.14.3. Eosael Sensor Performance Model -- 2.15. Summary -- References. Chapter 3. Data Fusion Algorithms And Architectures -- 3.1. Definition Of Data Fusion -- 3.2. Level 1 Processing -- 3.3. Level 2, 3, And 4 Processing -- 3.4. Data Fusion Processor Functions -- 3.5. Definition Of An Architecture -- 3.6. Data Fusion Architectures -- 3.7. Sensor Footprint Registration And Size Considerations -- 3.8. Summary -- References. Chapter 4. Classical Inference -- 4.1. Estimating The Statistics Of A Population -- 4.2. Interpreting The Confidence Interval -- 4.3. Confidence Interval For A Population Mean -- 4.4. Significance Tests For Hypotheses -- 4.5. The Z-test For The Population Mean -- 4.6. Tests With Fixed Significance Level -- 4.7. The T-test For A Population Mean -- 4.8. Caution In Use Of Significance Tests -- 4.9. Inference As A Decision -- 4.10. Summary -- References. Chapter 5. Bayesian Inference -- 5.1. Bayes' Rule -- 5.2. Bayes' Rule In Terms Of Odds Probability And Likelihood Ratio -- 5.3. Direct Application Of Bayes' Rule To Cancer Screening Test Example -- 5.4. Comparison Of Bayesian Inference With Classical Inference -- 5.5. Application Of Bayesian Inference To Fusing Information From Multiple Sources -- 5.6. Combining Multiple Sensor Information Using The Odds Probability Form Of Bayes' Rule -- 5.7. Recursive Bayesian Updating -- 5.8. Posterior Calculation Using Multivalued Hypotheses And Recursive Updating -- 5.9. Enhancing Underground Mine Detection With Data From Two Noncommensurate Sensors -- 5.10. Summary -- References. Chapter 6. Dempster-shafer Evidential Theory -- 6.1. Overview Of The Process -- 6.2. Implementation Of The Method -- 6.3. Support, Plausibility, And Uncertainty Interval -- 6.4. Dempster's Rule For Combination Of Multiple Sensor Data -- 6.5. Comparison Of Dempster-shafer With Bayesian Decision Theory -- 6.6 Probabilistic Models For Transformation Of Dempster-shafer Belief Functions For Decision Making -- 6.7. Summary -- References. Chapter 7. Artificial Neural Networks -- 7.1. Applications Of Artificial Neural Networks -- 7.2. Adaptive Linear Combiner -- 7.3. Linear Classifiers -- 7.4. Capacity Of Linear Classifiers -- 7.5. Nonlinear Classifiers -- 7.6. Capacity Of Nonlinear Classifiers -- 7.7. Supervised And Unsupervised Learning -- 7.8. Supervised Learning Rules -- 7.9. Generalization -- 7.10. Other Artificial Neural Networks And Processing Techniques -- 7.11. Summary -- References. Chapter 8. Voting Logic Fusion -- 8.1. Sensor Target Reports -- 8.2. Sensor Detection Space -- 8.3. System Detection Probability -- 8.4. Application Example Without Singleton Sensor Detection Modes -- 8.5. Hardware Implementation Of Voting Logic Sensor Fusion -- 8.6. Application Example With Singleton Sensor Detection Modes -- 8.7. Comparison Of Voting Logic Fusion With Dempster-shafer Evidential Theory -- 8.8. Summary -- References. Chapter 9. Fuzzy Logic And Fuzzy Neural Networks -- 9.1. Conditions Under Which Fuzzy Logic Provides An Appropriate Solution -- 9.2. Illustration Of Fuzzy Logic In An Automobile Antilock System -- 9.3. Basic Elements Of A Fuzzy System -- 9.4. Fuzzy Logic Processing -- 9.5. Fuzzy Centroid Calculation -- 9.6. Balancing An Inverted Pendulum With Fuzzy Logic Control -- 9.7. Fuzzy Logic Applied To Multitarget Tracking -- 9.8. Fuzzy Neural Networks -- 9.9. Fusion Of Fuzzy-valued Information From Multiple -- Sources -- 9.10. Summary -- References. Chapter 10. Passive Data Association Techniques For Unambiguous Location Of Targets -- 10.1. Data Fusion Options -- 10.2. Received-signal Fusion -- 10.3. Angle Data Fusion -- 10.4. Decentralized Fusion Architecture -- 10.5. Passive Computation Of Range Using Tracks From A Single Sensor Site -- 10.6. Summary -- References. Chapter 11. Retrospective Comments -- Appendix A. Planck Radiation Law And Radiative Transfer -- A.1. Planck Radiation Law -- A.2. Radiative Transfer Theory -- References -- Appendix B. Voting Fusion With Nested Confidence Levels -- Index. Lawrence A. Klein. Includes Bibliographical References And Index.

This book illustrates the benefits of sensor fusion by considering the characteristics of infrared, microwave, and millimeter-wave sensors, including the influence of the atmosphere on their performance. Applications that benefit from this technology include: vehicular traffic management, remote sensing, target classification and tracking- weather forecasting- military and homeland defense. Covering data fusion algorithms in detail, Klein includes a summary of the information required to implement each of the algorithms discussed, and outlines system application scenarios that may limit sensor size but that require high resolution data.--Résumé de l'éditeur

This book illustrates the benefits of sensor fusion by considering the characteristics of infrared, microwave, and millimeter-wave sensors, including the influence of the atmosphere on their performance. Applications that benefit from this technology include: vehicular traffic management, remote sensing, target classification and tracking, weather forecasting, military and homeland defense. Covering data fusion algorithms in detail, the author includes a summary of the information required to implement each of the algorithms discussed, and outlines system application scenarios that may limit sensor size but that require high resolution data.

This book illustrates the benefits of sensor fusion by considering the characteristics of infrared, microwave, and millimeter-wave sensors, including the influence of the atmosphere on their performance. Topics include applications of multiple-sensor systems; target, background, and atmospheric signature-generation phenomena and modeling; and methods of combining multiple-sensor data in target identity and tracking data fusion architectures. The information in this edition has been substantially expanded and updated to incorporate recent approaches to sensor and data fusion and application examples.

Content:
Front Matter
• List of Figures
• List of Tables
• Preface
• Table of Contents
• 1. Introduction
2. Multiple Sensor System Applications, Benefits, and Design Considerations
3. Data Fusion Algorithms and Architectures
4. Classical Inference
5. Bayesian Inference
6. Dempster-Shafer Evidential Theory
7. Artificial Neural Networks
8. Voting Logic Fusion
9. Fuzzy Logic and Fuzzy Neural Networks
10. Passive Data Association Techniques for Unambiguous Location of Targets
• 11. Retrospective Comments
Appendices
Index
• About the Author

2013-08-14