Foundations and Applications of Sensor Management (eBook)

eBook Download: PDF
2007 | 2008
XVIII, 310 Seiten
Springer US (Verlag)
978-0-387-49819-5 (ISBN)

Lese- und Medienproben

Foundations and Applications of Sensor Management -
Systemvoraussetzungen
149,79 inkl. MwSt
  • Download sofort lieferbar
  • Zahlungsarten anzeigen

This book covers control theory signal processing and relevant applications in a unified manner. It introduces the area, takes stock of advances, and describes open problems and challenges in order to advance the field. The editors and contributors to this book are pioneers in the area of active sensing and sensor management, and represent the diverse communities that are targeted.


Foundations and Applications of Sensor Management presents the emerging theory of sensor management with applications to real-world examples such as landmine detection, adaptive signal and image sampling, multi-target tracking, and radar waveform scheduling. It is written by leading experts in the field for a diverse engineering audience ranging from signal processing, to automatic control, statistics, and machine learning. The level of treatment of the book is tutorial and self-contained. The chapters of the book follow a logical development from theoretical foundations to approximate approaches and ending with applications. The coverage includes the following topics: stochastic control foundations of sensor management; multi-armed bandits and their connections to sensor management; information-theoretic approaches; managed sensing for multi-target tracking; approximation methods based on embedded simulation; active learning for classification and sampling; and waveform scheduling for radar. An appendix is included to provide essential background on topics the reader may not have encountered as a first-year graduate student: Markov decision processes; information theory; and stopping times.Foundations and Applications of Sensor Management is an important reference for signal processing and control engineers and researchers as well as machine learning application developers.

Preface 6
Acknowledgments 7
Contents 8
Contributing Authors 13
Symbol Index 15
OVERVIEW OF BOOK 17
1. Introduction 17
2. Scope of Book 18
3. Book Organization 19
STOCHASTIC CONTROL THEORY FOR SENSOR MANAGEMENT 22
1. Introduction 22
2. Markov Decision Problems 25
2.1 Dynamic Programming 27
2.2 Stationary Problems 28
2.3 Algorithms for MDPs 33
3. Partially Observed Markov Decision Problems 34
3.1 MDP Representation of POMDPs 36
3.2 Dynamic Programming for POMDPs 39
4. Approximate Dynamic Programming 41
5. Example 42
6. Conclusion 47
INFORMATION THEORETIC APPROACHES TO SENSOR MANAGEMENT 48
1. Introduction 48
2. Background 50
2.1 a-Entropy, a-Conditional Entropy, and a-Divergence 51
2.2 Relations Between Information Divergence and Risk 53
2.3 Fisher Information and Information Divergence 55
3. Information-Optimal Policy Search 55
4. Information Gain Via Classification Reduction 58
5. A Near Universal Proxy 59
6. Information Theoretic Sensor Management for Multi- target Tracking 62
6.1 The Model Multi-target Tracking Problem 63
6.2 R ´ enyi Divergence for Sensor Scheduling 64
6.3 Multi-target Tracking Experiment 65
6.4 On the Choice of 65
6.5 Sensitivity to Model Mismatch 66
6.6 Information Gain vs Entropy Reduction 67
7. Terrain Classification in Hyperspectral Satellite Imagery 68
7.1 Optimal Waveform Selection 69
8. Conclusion and Perspectives 72
JOINT MULTI-TARGET PARTICLE FILTERING 73
1. Introduction 73
2. The Joint Multi-target Probability Density 76
2.1 General Bayesian Filtering 78
2.2 Non-Linear Bayesian Filtering for a Single Target 79
2.3 Accounting for Target Birth and Death 80
2.4 Computing Renyi Divergence 81
2.5 Sensor Modeling 82
3. Particle Filter Implementation of JMPD 85
3.1 The Single Target Particle Filter 86
3.2 The Multi-target Particle Filter 87
3.3 Permutation Symmetry and Improved Importance Densities for JMPD 88
3.4 Multi-target Particle Proposal Via Individual Target Proposals 89
3.5 Multi-target Particle Proposal Via Joint Sampling 93
3.6 Partition Ordering 95
3.7 Estimation 97
3.8 Resampling 99
4. Multi-target Tracking Experiments 99
4.1 Adaptive Proposal Results 100
4.2 Partition Swapping 103
4.3 The Value of Not Thresholding 103
4.4 Unknown Number of Targets 104
5. Conclusions 105
POMDP APPROXIMATION USING SIMULATION AND HEURISTICS 108
1. Introduction 108
2. Motivating Example 110
3. Basic Principle: Q-value Approximation 111
3.1 Optimal Policy 111
3.2 Q-values 112
3.3 Stationary policies 113
3.4 Receding horizon 113
3.5 Approximating Q-values 113
4. Control Architecture 114
4.1 Controller 115
4.2 Measurement filter 115
4.3 Action selector 116
5. Q-value Approximation Methods 117
5.1 Basic approach 117
5.2 Monte Carlo sampling 117
5.3 Relaxation of optimization problem 118
5.4 Heuristic approximation 119
5.5 Parametric approximation 120
5.6 Action-sequence approximations 122
5.7 Rollout 123
5.8 Parallel rollout 124
5.9 Control architecture in the Monte Carlo case 124
5.10 Belief-state simplification 127
5.11 Reward surrogation 128
6. Simulation Result 129
7. Summary and Discussion 131
MULTI-ARMED BANDIT PROBLEMS 133
1. Introduction 133
2. The Classical Multi-armed Bandit 134
2.1 Problem Formulation 135
2.2 On Forward Induction 137
2.3 Key Features of the Classical MAB Problem and the Nature of its Solution 139
2.4 Computational Issues 142
3. Variants of the Multi-armed Bandit Problem 146
3.1 Superprocesses 146
3.2 Arm-acquiring Bandits 149
3.3 Switching Penalties 150
3.4 Multiple Plays 152
3.5 Restless Bandits 154
3.6 Discussion 159
4. Example 160
5. Chapter Summary 163
APPLICATION OF MULTI-ARMED BANDITS TO SENSOR MANAGEMENT 164
1. Motivating Application and Overview 164
1.1 Introduction 164
1.2 SM Example of Multi-armed Bandit 165
1.3 Organization and Notation of This Chapter 166
2. Application to Sensor Management 166
2.1 Application of the Classical MAB 167
2.2 Single Sensor with Multiple Modes 170
2.3 Detecting New Targets 171
2.4 Sensor Switching Delays 171
2.5 Multiple Sensors 172
2.6 Application of Restless Bandits 173
3. Example Application 173
3.1 MAB Formulation of SM Tracking Problem 174
3.2 Index Rule Solution of MAB 175
3.3 Numerical Results and Comparison to Other Solutions 177
4. Summary and Discussion 184
ACTIVE LEARNING AND SAMPLING 187
1. Introduction 187
1.1 Some Motivation Examples 188
2. A Simple One-dimensional Problem 189
3. Beyond 1d - Piecewise Constant Function Estimation 200
4. Final Remarks and Open Questions 209
PLAN-IN-ADVANCE ACTIVE LEARNING OF CLASSIFIERS 211
1. Introduction 211
2. Analytical Forms of the Classifier 213
3. Pre-labeling Selection of Basis Functions 214
4. Pre-labeling Selection of Data 219
5. Connection to Theory of Optimal Experiments 220
6. Application to UXO Detection 222
6.1 Magnetometer and electromagnetic induction sensors 223
6.2 Measured sensor data from the Jefferson Proving Ground 223
6.3 Detection results 224
7. Chapter Summary 229
APPLICATION OF SENSOR SCHEDULING CONCEPTS TO RADAR 231
1. Introduction 231
2. Basic Radar 232
2.1 Range-Doppler Ambiguity 232
2.2 Elevation and Azimuth 236
2.3 Doppler Processing 239
2.4 Effects of Ambiguity 242
3. Measurement in Radar 243
4. Basic Scheduling of Waveforms in Target Tracking 244
4.1 Measurement Validation 245
4.2 IPDA Tracker 245
5. Measures of Effectiveness for Waveforms 249
5.1 Single Noise Covariance Model 250
5.2 Integrated Clutter Measure 250
5.3 Approximation of ICM 251
5.4 Simulation Results 253
6. Scheduling of Beam Steering and Waveforms 255
6.1 Tracking of Multiple Maneuvering Targets 255
6.2 Scheduling 256
6.3 Simulation results 258
7. Waveform Libraries 260
7.1 LFM Waveform Library 263
7.2 LFM-Rotation Library 264
8. Conclusion 265
DEFENSE APPLICATIONS 267
1. Introduction 267
2. Background 269
3. The Contemporary Situation 270
4. Dynamic Tactical Targeting (DTT) 272
5. Conclusion 276
APPENDICES 279
1. Information Theory 279
1.1 Entropy and Conditional Entropy 279
1.2 Information Divergence 281
1.3 Shannon’s Data Processing Theorem 281
1.4 Shannon Mutual Information 282
1.5 Further Reading 282
2. Markov Processes 283
2.1 Definition of Markov Process 283
2.2 State-transition Probability 284
2.3 Chapman-Kolmogorov Equation 285
2.4 Markov reward processes 285
2.5 Partially Observable Markov Processes 286
2.6 Further Reading 288
3. Stopping Times 288
3.1 Definitions 288
3.2 Example 289
3.3 Stopping Times for Multi-armed Bandit Problems 290
3.4 Further Reading 291
References 292
Index 313

Erscheint lt. Verlag 23.10.2007
Reihe/Serie Signals and Communication Technology
Signals and Communication Technology
Zusatzinfo XVIII, 310 p.
Verlagsort New York
Sprache englisch
Themenwelt Mathematik / Informatik Informatik
Technik Elektrotechnik / Energietechnik
Technik Nachrichtentechnik
Schlagworte active sensing • automatic controls • Castañón • Cochran • Filter • Hero • Information • Information Theory • Kalman Filtering • Kastella • machine learning • Radar • Sensor • target detection • Tracking • tracking identification • waveform • waveform sensor
ISBN-10 0-387-49819-2 / 0387498192
ISBN-13 978-0-387-49819-5 / 9780387498195
Haben Sie eine Frage zum Produkt?
PDFPDF (Wasserzeichen)
Größe: 7,2 MB

DRM: Digitales Wasserzeichen
Dieses eBook enthält ein digitales Wasser­zeichen und ist damit für Sie persona­lisiert. Bei einer missbräuch­lichen Weiter­gabe des eBooks an Dritte ist eine Rück­ver­folgung an die Quelle möglich.

Dateiformat: PDF (Portable Document Format)
Mit einem festen Seiten­layout eignet sich die PDF besonders für Fach­bücher mit Spalten, Tabellen und Abbild­ungen. Eine PDF kann auf fast allen Geräten ange­zeigt werden, ist aber für kleine Displays (Smart­phone, eReader) nur einge­schränkt geeignet.

Systemvoraussetzungen:
PC/Mac: Mit einem PC oder Mac können Sie dieses eBook lesen. Sie benötigen dafür einen PDF-Viewer - z.B. den Adobe Reader oder Adobe Digital Editions.
eReader: Dieses eBook kann mit (fast) allen eBook-Readern gelesen werden. Mit dem amazon-Kindle ist es aber nicht kompatibel.
Smartphone/Tablet: Egal ob Apple oder Android, dieses eBook können Sie lesen. Sie benötigen dafür einen PDF-Viewer - z.B. die kostenlose Adobe Digital Editions-App.

Buying eBooks from abroad
For tax law reasons we can sell eBooks just within Germany and Switzerland. Regrettably we cannot fulfill eBook-orders from other countries.

Mehr entdecken
aus dem Bereich
Konzepte, Methoden, Lösungen und Arbeitshilfen für die Praxis

von Ernst Tiemeyer

eBook Download (2023)
Carl Hanser Verlag GmbH & Co. KG
69,99
Konzepte, Methoden, Lösungen und Arbeitshilfen für die Praxis

von Ernst Tiemeyer

eBook Download (2023)
Carl Hanser Verlag GmbH & Co. KG
69,99
Der Weg zur professionellen Vektorgrafik

von Uwe Schöler

eBook Download (2024)
Carl Hanser Verlag GmbH & Co. KG
29,99