Muutke küpsiste eelistusi

Compressive Sensing for Urban Radar [Pehme köide]

Edited by (Villanova University, Pennsylvania, USA)
  • Formaat: Paperback / softback, 508 pages, kõrgus x laius: 234x156 mm, kaal: 940 g, 9 Tables, black and white; 196 Illustrations, black and white
  • Ilmumisaeg: 29-Mar-2017
  • Kirjastus: CRC Press
  • ISBN-10: 1138073407
  • ISBN-13: 9781138073401
Teised raamatud teemal:
Compressive Sensing for Urban RadarSuurem pilt
  • Formaat: Paperback / softback, 508 pages, kõrgus x laius: 234x156 mm, kaal: 940 g, 9 Tables, black and white; 196 Illustrations, black and white
  • Ilmumisaeg: 29-Mar-2017
  • Kirjastus: CRC Press
  • ISBN-10: 1138073407
  • ISBN-13: 9781138073401
Teised raamatud teemal:
Püsilink: https://www.kriso.ee/db/9781138073401.html
With the emergence of compressive sensing and sparse signal reconstruction, approaches to urban radar have shifted toward relaxed constraints on signal sampling schemes in time and space, and to effectively address logistic difficulties in data acquisition. Traditionally, these challenges have hindered high resolution imaging by restricting both bandwidth and aperture, and by imposing uniformity and bounds on sampling rates.Compressive Sensing for Urban Radar is the first book to focus on a hybrid of two key areas: compressive sensing and urban sensing. It explains how reliable imaging, tracking, and localization of indoor targets can be achieved using compressed observations that amount to a tiny percentage of the entire data volume. Capturing the latest and most important advances in the field, this state-of-the-art text:Covers both ground-based and airborne synthetic aperture radar (SAR) and uses different signal waveformsDemonstrates successful applications of compressive sensing for target detection and revealing building interiors Describes problems facing urban radar and highlights sparse reconstruction techniques applicable to urban environmentsDeals with both stationary and moving indoor targets in the presence of wall clutter and multipath exploitationProvides numerous supporting examples using real data and computational electromagnetic modeling Featuring 13 chapters written by leading researchers and experts, Compressive Sensing for Urban Radar is a useful and authoritative reference for radar engineers and defense contractors, as well as a seminal work for graduate students and academia.

Arvustused

"The essential feature of this book is that it brings together the areas of compressive sensing and radar imaging for urban sensing. These areas of attributes are highly relevant to promote sustainability and for a range of civil and military applications, such as search and rescue missions, hostage rescue situations, urban design, and surveillance and reconnaissance in urban environments." Fulvio Gini, University of Pisa, Italy

Preface vii
Editor xi
Contributors xiii
1 Compressive Sensing Fundamentals 1(48)
Michael B. Wakin
1.1 Overview
2(5)
1.1.1 Signal Models and Dimensionality Reduction
2(1)
1.1.2 Motivation for Compressive Sensing
3(2)
1.1.3 Compressive Sensing in a Nutshell
5(2)
1.2 Sparse Modeling
7(8)
1.2.1 Sparsity, Compressibility, and Norms
7(1)
1.2.2 Sparsity in Orthonormal Bases
8(2)
1.2.3 Sparsity in Nonorthonormal Dictionaries
10(2)
1.2.3.1 Synthesis Sparsity
10(1)
1.2.3.2 Analysis Sparsity
11(1)
1.2.4 Extensions of Sparse Models
12(3)
1.2.4.1 Structured Sparsity Models
12(1)
1.2.4.2 Statistical Sparsity Models
13(1)
1.2.4.3 Beyond Sparsity
14(1)
1.3 Compressive Measurement Protocols
15(9)
1.3.1 Random Gaussian and Subgaussian Matrices
16(3)
1.3.2 Random Sampling in an Orthogonal Basis
19(1)
1.3.3 Measurement Systems
20(4)
1.3.3.1 One-Dimensional Signals
21(1)
1.3.3.2 Images and Higher-Dimensional Signals
22(2)
1.4 Sparse Signal Recovery Algorithms and Guarantees
24(12)
1.4.1 Preliminaries
24(2)
1.4.2 Optimization-Based Recovery from Noise-Free Measurements
26(2)
1.4.2.1 Problem Formulation
26(1)
1.4.2.2 Performance Guarantees
26(2)
1.4.2.3 Computational Considerations
28(1)
1.4.3 Optimization-Based Recovery from Noisy Measurements
28(2)
1.4.3.1 Problem Formulation
28(1)
1.4.3.2 Performance Guarantees
29(1)
1.4.3.3 Computational Considerations
29(1)
1.4.3.4 Parameter Selection
30(1)
1.4.4 Greedy Methods
30(4)
1.4.4.1 Orthogonal Matching Pursuit (OMP)
30(2)
1.4.4.2 Compressive Sampling Matching Pursuit (CoSaMP)
32(1)
1.4.4.3 Iterative Hard Thresholding (IHT)
33(1)
1.4.5 Signal Recovery in Nonorthonormal Dictionaries
34(16)
1.4.5.1 Synthesis Sparsity in Redundant Dictionaries
34(1)
1.4.5.2 Analysis Sparsity
35(1)
Acknowledgments
36(1)
References
37(12)
2 Overcomplete Dictionary Design for Building Feature Extraction 49(38)
Wim van Rossum
Jacco de Wit
2.1 Introduction
50(5)
2.1.1 Overview of Through-Wall Radar Mapping
50(2)
2.1.2 Typical Measurement Geometry
52(1)
2.1.3 Bases, Frames, and Overcomplete Dictionaries
52(3)
2.1.4 Layout of This
Chapter
55(1)
2.2 Building Feature Extraction
55(4)
2.2.1 Point Scattering Focusing
55(2)
2.2.2 Smashed Filter Processing
57(1)
2.2.3 OCD with Sparse Representation
58(1)
2.3 How to Create an OCD
59(3)
2.3.1 Knowledge-Based Dictionaries
59(1)
2.3.2 Adaptive Knowledge-Based Dictionaries
60(1)
2.3.3 Learned Dictionary
61(1)
2.4 Practical Atom Definition
62(7)
2.4.1 Starting Points
62(3)
2.4.2 Atom Definition
65(4)
2.5 Through-Wall Radar Measurements
69(13)
2.5.1 Building Layout
70(1)
2.5.2 Point Scattering Focusing
71(3)
2.5.2.1 Reflectivity Maps
71(1)
2.5.2.2 Classified Scatterers
72(2)
2.5.3 Smashed Filter Processing
74(4)
2.5.3.1 Reflectivity Maps
74(2)
2.5.3.2 Classified Scatterers
76(2)
2.5.4 OCD with Sparse Representation
78(16)
2.5.4.1 CAMP Algorithm
79(1)
2.5.4.2 Reflectivity Maps
80(1)
2.5.4.3 Classified Scatterers
80(2)
2.6 Conclusion
82(2)
References
84(3)
3 Compressive Sensing for Radar Imaging of Underground Targets 87(36)
Kyle R. Krueger
James H. McClellan
Waymond R. Scott Jr
3.1 Introduction
88(2)
3.2 Background for GPR Imaging
90(4)
3.3 System Framework with CS
94(11)
3.3.1 Designing Phi
95(4)
3.3.1.1 Time Pulse Phi
96(2)
3.3.1.2 Stepped Frequency Phi
98(1)
3.3.2 CS Inversion
99(2)
3.3.3 Compressed Orthogonal Matching Pursuit
101(1)
3.3.4 Basic CS Simulation
102(3)
3.4 Computational Reductions
105(10)
3.4.1 Shift Invariance Property
106(4)
3.4.2 Implementation Specifics for Structure Change
110(3)
3.4.3 Simulation Using Functional Dictionary
113(2)
3.5 Applied Performance: Laboratory Data
115(4)
3.5.1 Air-Target Experiment
116(1)
3.5.2 Subsurface-Target Experiment
117(2)
3.6 Conclusions
119(1)
References
119(4)
4 Wall Clutter Mitigations for Compressive Imaging of Building Interiors 123(30)
Fauzia Ahmad
4.1 Introduction
124(3)
4.2 Wall Mitigation Techniques of Spatial Filtering and Subspace Projection
127(5)
4.2.1 Through-the-Wall Signal Model
127(2)
4.2.2 Wall Clutter Mitigation Techniques
129(2)
4.2.2.1 Spatial Filtering
129(1)
4.2.2.2 Subspace Projection
130(1)
4.2.3 Scene Reconstruction
131(1)
4.2.4 Illustrative Results
132(1)
4.3 Spatial Filtering and Subspace Projection under Reduced Data Volume
132(4)
4.3.1 Wall Mitigations under Reduced Data Volume
134(1)
4.3.2 CS-Based Scene Reconstruction
134(1)
4.3.3 Illustrative Results
135(1)
4.4 Wall Clutter Mitigation Using DPSSs
136(5)
4.4.1 Discrete Prolate Spheroidal Sequences
137(1)
4.4.2 DPSS Basis
137(2)
4.4.3 Block-Sparse Reconstruction
139(1)
4.4.4 Illustrative Results
140(1)
4.5 Partially Sparse Reconstruction of Indoor Scenes
141(7)
4.5.1 Partially Sparse Signal Model
142(2)
4.5.2 Sparse Scene Reconstruction
144(1)
4.5.3 Illustrative Results
145(3)
4.6 Conclusion
148(1)
References
148(5)
5 Compressive Sensing for Urban Multipath Exploitation 153(44)
Michael Leigsnering
Abdelhak M. Zoubir
5.1 Introduction
154(1)
5.2 Ultrawideband Signal Model
155(5)
5.2.1 Relation to the Stationary Scene Model
158(1)
5.2.2 Conventional Image Formation
159(1)
5.3 Multipath Propagation Model
160(10)
5.3.1 Interior Wall Multipath
162(1)
5.3.2 Wall Ringing Multipath
163(2)
5.3.3 Bistatic Received Signal Model
165(5)
5.4 Compressive Sensing Reconstruction with Multipath Exploitation
170(7)
5.4.1 Stationary Scenes
170(1)
5.4.2 Group Sparse Reconstruction of Stationary Scenes
170(3)
5.4.3 Example
173(4)
5.4.3.1 Simulation Results
174(1)
5.4.3.2 Experimental Results
175(2)
5.4.4 Moving Targets
177(5)
5.4.5 Group Sparse Reconstruction of Stationary/Nonstationary Scenes
178(1)
5.4.6 Example
179(3)
5.4.6.1 Simulation Results
179(2)
5.4.6.2 Experimental Results
181(1)
5.5 Compressive Sensing Reconstruction with the Wall Included
182(8)
5.5.1 Wall Reverberation Model
184(2)
5.5.2 Separate Reconstruction
186(1)
5.5.3 Joint Group Sparse Reconstruction
186(2)
5.5.4 Example
188(12)
5.5.4.1 Simulation Results
188(2)
5.5.4.2 Experimental Results
190(1)
5.6 Conclusion
190(2)
Acknowledgments
192(1)
References
192(5)
6 Measurement Kernel Design for HRR Imaging of Urban Objects 197(34)
Nathan A. Goodman
Yujie Gu
Junhyeong Bae
6.1 Introduction
198(2)
6.2 Sub-Nyquist Sampling Implementations, Models, and Constraints
200(8)
6.2.1 Implementation of Sub-Nyquist Sampling
201(2)
6.2.2 Power and Cost Benefits
203(1)
6.2.3 Measurement Model with Preprojection Additive Noise
204(2)
6.2.4 Matrix-Vector Measurement Model
206(2)
6.3 Radar Target and Received Signal Models
208(3)
6.3.1 Linear Target Model
208(3)
6.3.2 Compression Ratio
211(1)
6.3.3 Signal-to-Noise Ratio
211(1)
6.4 Information-Based Measurement Kernel Optimization
211(8)
6.4.1 Task-Specific Information
212(1)
6.4.2 Approximate Gradient of TSI for a Gaussian Mixture Model
213(4)
6.4.3 Application to HRR Imaging
217(2)
6.4.4 MMSE HRR Estimation
219(1)
6.5 Simulation Results
219(8)
6.5.1 Training Data and Gaussian Mixture Calculation
219(1)
6.5.2 Waveforms and Compression Ratio
220(1)
6.5.3 Signal Examples
221(2)
6.5.4 Quantitative Performance Results
223(4)
6.6 Conclusions
227(1)
Acknowledgments
228(1)
References
228(3)
7 Compressive Sensing for Multipolarization through-the-Wall Radar Imaging 231(20)
Abdesselam Bouzerdoum
Jack Yang
Fok Hing Chi Tivive
7.1 Introduction
232(1)
7.2 Through-the-Wall Radar Imaging
233(5)
7.2.1 Delay-and-Sum Beamforming
233(2)
7.2.2 Single-Polarization Imaging Using SMV Model
235(2)
7.2.3 Multipolarization Imaging Using SMV Model
237(1)
7.3 Multipolarization Imaging Using MMV Model
238(2)
7.3.1 MMV CS Model
238(1)
7.3.2 Joint Image Fusion and Formation Using MMV
238(2)
7.4 Experimental Results
240(6)
7.4.1 Experimental Results Using Synthetic Data
241(2)
7.4.2 Experimental Results Using Real Data
243(3)
7.5 Conclusion
246(2)
References
248(3)
8 Sparsity-Aware Human Motion Indication 251(32)
Moeness G. Amin
8.1 Introduction
252(2)
8.2 Change Detection
254(13)
8.2.1 Backprojection-Based Change Detection
254(2)
8.2.2 Sparsity-Driven Change Detection under Translational Motion
256(2)
8.2.3 Sparsity-Driven Change Detection under Short Sudden Movements
258(4)
8.2.4 Experimental Results for Change Detection
262(5)
8.3 Sparsity for Target Localization and Motion Parameter Estimation
267(10)
8.3.1 UWB Signal Model
268(3)
8.3.2 Backprojection-Based Stationary and Moving Target Localization
271(2)
8.3.3 CS-Based Stationary and Moving Target Localization
273(3)
8.3.3.1 Linear Model Formulation
273(1)
8.3.3.2 CS Data Acquisition and Scene Reconstruction
274(2)
8.3.4 Experimental Results
276(1)
8.4 Conclusions
277(2)
References
279(4)
9 Time-Frequency Analysis of Micro-Doppler Signals Based on Compressive Sensing 283(44)
Ljubisa Stankovic
Srdjan Stankovic
Irena Orovic
Yimin D. Zhang
9.1 Introduction
284(2)
9.2 Background
286(3)
9.2.1 Time-Varying Micro-Doppler Signatures
286(3)
9.2.2 Human Gait Modeling
289(1)
9.3 Time-Frequency Analysis of the m-D and Rigid Body Signals
289(6)
9.3.1 Missing STFT Samples due to the m-D Removal
291(4)
9.4 Sparse Compressive Sensed Signals and Time-Frequency Analysis
295(10)
9.4.1 Missing Samples due to Reduced Sampling Rate
295(3)
9.4.2 Analysis of Missing Samples in the FT (STFT) Domain
298(4)
9.4.3 Effects of Missing Samples to Bilinear Time-Frequency Distributions
302(3)
9.5 CS Reconstructions in Time-Frequency Domain
305(18)
9.5.1 Signal Reconstruction from Bilinear Transforms
305(6)
9.5.1.1 Ambiguity Domain-Based CS
305(2)
9.5.1.2 IAF Reconstruction Yielding Sparse TFR
307(1)
9.5.1.3 Robust Ambiguity Domain-Based CS in the Presence of Impulse Noise
308(3)
9.5.2 Compressive Sensing Reconstruction from Linear Transforms
311(7)
9.5.2.1 Reconstruction Based on the CS Methods
314(4)
9.5.3 Windows with Overlapping
318(5)
9.6 Conclusion
323(1)
References
324(3)
10 Urban Target Tracking Using Sparse Representations 327(34)
Phani Chavali
Arye Nehorai
10.1 Introduction
328(2)
10.2 System Model
330(6)
10.2.1 Multipath Environment Model
330(4)
10.2.2 Signal Model
334(1)
10.2.3 State-Space Model
335(1)
10.2.4 Measurement Model
336(1)
10.3 Sparse Modeling
336(6)
10.4 Sparsity-Based Multiple-Target Tracking
342(9)
10.4.1 Standard Sparse Signal Reconstruction Techniques
343(1)
10.4.2 Effect of Multipath Environment
344(5)
10.4.3 PB Support Recovery Algorithm
349(2)
10.5 Numerical Results
351(3)
10.6 Conclusions
354(2)
References
356(5)
11 Three-Dimensional Imaging of Vehicles from Sparse Apertures in Urban Environment 361(26)
Emre Ertin
11.1 Introduction
362(1)
11.2 System Model
363(4)
11.3 Case Study for 3-D SAR: AFRL GOTCHA Volumetric SAR Data Set
367(2)
11.4 Direct Approach to Sparsity-Regularized 3-D Construction
369(4)
11.4.1 Algorithmic and Computational Considerations
371(2)
11.5 Multiple Elevation IFSAR
373(8)
11.5.1 Sparsity-Regularized Interpolation Approach to m-IFSAR
376(3)
11.5.2 DFT Peak Detection Approach for m-IFSAR
379(2)
11.6 Practical Considerations: Autofocus and Registration
381(2)
References
383(4)
12 Compressive Sensing for MIMO Urban Radar 387(42)
Yao Yu
Athina Petropulu
Rabinder N. Madan
12.1 Introduction
388(1)
12.1.1 Outline of the
Chapter
389(1)
12.2 Colocated CS-MIMO Radars
389(8)
12.2.1 Problem Formulation and Solution
389(8)
12.3 Challenging Issues Associated with CS-MIMO Radars
397(7)
12.3.1 Basis Mismatch and Resolution
398(1)
12.3.2 Complexity
398(1)
12.3.3 Clutter Rejection: CS-Capon
398(6)
12.3.4 Phase Synchronization
404(1)
12.4 Advanced Techniques for CS-MIMO Radars
404(14)
12.4.1 Power Allocation
404(7)
12.4.2 Waveform Design for Colocated CS-MIMO Radars
411(1)
12.4.3 Measurement Matrix Design
412(18)
12.4.3.1 Measurement Matrix Design by Reducing CSM and Increasing SIR
413(2)
12.4.3.2 Measurement Matrix Design by Improving SIR Only
415(2)
12.4.3.3 Phi#1 versus Phi#2
417(1)
12.5 Application to Through-the-Wall Radar
418(6)
Acknowledgments
424(1)
References
424(5)
13 Compressive Sensing Meets Noise Radar 429(32)
Mahesh C. Shastry
Ram M. Narayanan
Muralidhar Rangaswamy
13.1 Introduction
430(5)
13.1.1 State of the Art in Compressive Radar Imaging
433(2)
13.2 Basics of Compressive Stochastic Waveform Radar
435(7)
13.2.1 Compressive Radar
435(2)
13.2.2 Correlations in the Circulant Matrix
437(1)
13.2.3 Experiments
437(2)
13.2.4 Analysis of Experimental Data
439(1)
13.2.5 Imaging Performance
439(3)
13.3 Detection Strategies for Compressive Noise Radar
442(13)
13.3.1 Compressive Sensing Detection
442(3)
13.3.2 Statistics of the Error of Compressive Signal Recovery
445(3)
13.3.3 Threshold Estimation for Compressive Detection
448(2)
13.3.4 GPD and Compressive Sensing
450(1)
13.3.5 Computational Complexity of GPD-Based Threshold Estimation
451(4)
13.3.5.1 Performance of GPD-Based Threshold Estimation
452(3)
13.4 Conclusions and Future Work
455(2)
13.4.1 Compressive Noise Radar Imaging and Detection
455(1)
13.4.2 Open Problems
456(1)
References
457(4)
Index 461
Dr. Moeness G. Amin has been a faculty member of the Department of Electrical and Computer Engineering at Villanova University, Pennsylvania, USA for nearly 30 years. In 2002, he became the director of the Center for Advanced Communications, College of Engineering. Currently he is the chair of the Electrical Cluster of the Franklin Institute Committee on Science and the Arts, as well as an IEEE, SPIE, and IET fellow. The recipient of many prestigious awards, he has conducted extensive research in radar signal processing, authored over 650 journal and conference papers, and served as an editor for numerous publications.