Inverse Synthetic Aperture Radar Imaging With Matlab Algorithms 2Nd Edition Caner Ozdemir Full Chapter PDF

Inverse Synthetic Aperture Radar
Imaging With MATLAB Algorithms 2nd

Edition Caner Ozdemir
Visit to download the full and correct content document:
https://ebookmass.com/product/inverse-synthetic-aperture-radar-imaging-with-matlab
-algorithms-2nd-edition-caner-ozdemir/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Advanced Algorithms for Mineral and Hydrocarbon

Exploration Using Synthetic Aperture Radar 1st Edition
Maged Marghany
https://ebookmass.com/product/advanced-algorithms-for-mineral-
and-hydrocarbon-exploration-using-synthetic-aperture-radar-1st-
edition-maged-marghany/
Nonlinear Ocean Dynamics: Synthetic Aperture Radar 1st

Edition Maged Marghany
https://ebookmass.com/product/nonlinear-ocean-dynamics-synthetic-
aperture-radar-1st-edition-maged-marghany/
Introduction to Synthetic Aperture Radar: Concepts and

Practice E. David Jansing
https://ebookmass.com/product/introduction-to-synthetic-aperture-
radar-concepts-and-practice-e-david-jansing/
Modern Information Optics with MATLAB Yaping Zhang
https://ebookmass.com/product/modern-information-optics-with-
matlab-yaping-zhang/
Specialty Imaging: Arthrography 2nd Edition Julia Crim
Md
https://ebookmass.com/product/specialty-imaging-arthrography-2nd-
edition-julia-crim-md/
Fundamental Chemistry with MATLAB 1st Edition Daniele

Mazza
https://ebookmass.com/product/fundamental-chemistry-with-
matlab-1st-edition-daniele-mazza/
Signal processing for multistatic radar systems:

adaptive waveform selection, optimal geometries and
pseudolinear tracking algorithms Do■ançay
https://ebookmass.com/product/signal-processing-for-multistatic-
radar-systems-adaptive-waveform-selection-optimal-geometries-and-
pseudolinear-tracking-algorithms-dogancay/
Data Structures and Algorithms in C++ 2nd Edition
https://ebookmass.com/product/data-structures-and-algorithms-
in-c-2nd-edition/
Cryptography Algorithms - 2nd Edition (Early Release)

Massimo Bertaccini
https://ebookmass.com/product/cryptography-algorithms-2nd-
edition-early-release-massimo-bertaccini/
Inverse Synthetic Aperture
Radar Imaging with
MATLAB Algorithms
Wiley Series in Microwave and Optical Engineering
Kai Chang, Editor
Texas A&M University
A complete list of the titles in this series appears at the end of this volume.
Inverse Synthetic Aperture Radar Imaging
with MATLAB Algorithms
With Advanced SAR/ISAR Imaging Concepts,

Algorithms, and MATLAB Codes
Second Edition
Caner Özdemir, Phd

Mersin University
Mersin, Turkey
This second edition first published 2021
© 2021 John Wiley & Sons, Inc.
Edition History
John Wiley & Sons, Inc. (1e, 2012)
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or
transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise,
except as permitted by law. Advice on how to obtain permission to reuse material from this title is
available at http://www.wiley.com/go/permissions.
The right of Caner Özdemir to be identified as the author of this work has been asserted in accordance
with law.
Registered Office
John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA
Editorial Office
111 River Street, Hoboken, NJ 07030, USA
For details of our global editorial offices, customer services, and more information about Wiley
products visit us at www.wiley.com.
Wiley also publishes its books in a variety of electronic formats and by print-on-demand. Some
content that appears in standard print versions of this book may not be available in other formats.
Limit of Liability/Disclaimer of Warranty
MATLAB® is a trademark of The MathWorks, Inc. and is used with permission. The MathWorks
does not warrant the accuracy of the text or exercises in this book. This work’s use or discussion of
MATLAB® software or related products does not constitute endorsement or sponsorship by The
MathWorks of a particular pedagogical approach or particular use of the MATLAB® software. While
the publisher and authors have used their best efforts in preparing this work, they make no
representations or warranties with respect to the accuracy or completeness of the contents of this work
and specifically disclaim all warranties, including without limitation any implied warranties of
merchantability or fitness for a particular purpose. No warranty may be created or extended by sales
representatives, written sales materials or promotional statements for this work. The fact that an
organization, website, or product is referred to in this work as a citation and/or potential source of
further information does not mean that the publisher and authors endorse the information or services
the organization, website, or product may provide or recommendations it may make. This work is
sold with the understanding that the publisher is not engaged in rendering professional services. The
advice and strategies contained herein may not be suitable for your situation. You should consult
with a specialist where appropriate. Further, readers should be aware that websites listed in this work
may have changed or disappeared between when this work was written and when it is read. Neither
the publisher nor authors shall be liable for any loss of profit or any other commercial damages,
including but not limited to special, incidental, consequential, or other damages.
Library of Congress Cataloging-in-Publication Data
Names: Özdemir, Caner, author.
Title: Inverse synthetic aperture radar imaging with MATLAB algorithms:
with advanced sar/isar imaging concepts, algorithms, and matlab codes /
Caner Ozdemir, PhD Mersin University, Mersin, Turkey.
Description: 2nd edition. | Hoboken, NJ, USA : Wiley, 2021. | Series: Wiley
series in microwave and optical engineering | Includes bibliographical
references and index.
Identifiers: LCCN 2020031216 (print) | LCCN 2020031217 (ebook) | ISBN
9781119521334 (cloth) | ISBN 9781119521365 (adobe pdf) | ISBN
9781119521389 (epub)
Subjects: LCSH: MATLAB. | Synthetic aperture radar. | MATLAB.
Classification: LCC TK6592.S95 O93 2020 (print) | LCC TK6592.S95 (ebook)
| DDC 621.3848/5–dc23
LC record available at https://lccn.loc.gov/2020031216
LC ebook record available at https://lccn.loc.gov/2020031217
Cover design by Wiley
Cover image: Courtesy of Caner Özdemir, (background) © Maxiphoto/Getty Images
Set in 9.5/12.5pt STIXTwoText by SPi Global, Pondicherry, India
10 9 8 7 6 5 4 3 2 1
To:
My wife Betül,
My three daughters,
My brother,
My father,
and the memory of my beloved mother
vii
Contents
Preface to the Second Edition xvi

Acknowledgments xix
Acronyms xx
1 Basics of Fourier Analysis 1

1.1 Forward and Inverse Fourier Transform 1
1.1.1 Brief History of FT 1
1.1.2 Forward FT Operation 2
1.1.3 IFT 3
1.2 FT Rules and Pairs 3
1.2.1 Linearity 3
1.2.2 Time Shifting 3
1.2.3 Frequency Shifting 4
1.2.4 Scaling 4
1.2.5 Duality 4
1.2.6 Time Reversal 4
1.2.7 Conjugation 4
1.2.8 Multiplication 4
1.2.9 Convolution 5
1.2.10 Modulation 5
1.2.11 Derivation and Integration 5
1.2.12 Parseval’s Relationship 5
1.3 Time-Frequency Representation of a Signal 5
1.3.1 Signal in the Time Domain 6
1.3.2 Signal in the Frequency Domain 6
1.3.3 Signal in the Joint Time-Frequency (JTF) Plane 7
viii Contents
1.4 Convolution and Multiplication Using FT 11

1.5 Filtering/ Windowing 12
1.6 Data Sampling 14
1.7 DFT and FFT 16
1.7.1 DFT 16
1.7.2 FFT 17
1.7.3 Bandwidth and Resolutions 17
1.8 Aliasing 19
1.9 Importance of FT in Radar Imaging 19
1.10 Effect of Aliasing in Radar Imaging 23
1.11 Matlab Codes 26
References 33
2 Radar Fundamentals 35
2.1 Electromagnetic Scattering 35
2.2 Scattering from PECs 38
2.3 Radar Cross Section 39
2.3.1 Definition of RCS 40
2.3.2 RCS of Simple-Shaped Objects 43
2.3.3 RCS of Complex-Shaped Objects 44
2.4 Radar Range Equation 44
2.4.1 Bistatic Case 46
2.4.2 Monostatic Case 49
2.5 Range of Radar Detection 50
2.5.1 Signal-to-Noise Ratio 51
2.6 Radar Waveforms 53
2.6.1 Continuous Wave 53
2.6.2 Frequency-Modulated Continuous Wave 56
2.6.3 Stepped-Frequency Continuous Wave 59
2.6.4 Short Pulse 61
2.6.5 Chirp (LFM) Pulse 62
2.7 Pulsed Radar 69
2.7.1 Pulse Repetition Frequency 69
2.7.2 Maximum Range and Range Ambiguity 69
2.7.3 Doppler Frequency 70
2.8 Matlab Codes 74
References 82
Contents ix
3 Synthetic Aperture Radar 85

3.1 SAR Modes 86
3.2 SAR System Design 87
3.3 Resolutions in SAR 88
3.4 SAR Image Formation 91
3.5 Range Compression 92
3.5.1 Matched Filter 92
3.5.1.1 Computing Matched Filter Output via Fourier Processing 95
3.5.1.2 Example for Matched Filtering 96
3.5.2 Ambiguity Function 99
3.5.2.1 Relation to Matched Filter 100
3.5.2.2 Ideal Ambiguity Function 101
3.5.2.3 Rectangular-Pulse Ambiguity Function 102
3.5.2.4 LFM-Pulse Ambiguity Function 102
3.5.3 Pulse Compression 105
3.5.3.1 Detailed Processing of Pulse Compression 105
3.5.3.2 Bandwidth, Resolution, and Compression Issues for LFM Signal 109
3.5.3.3 Pulse Compression Example 110
3.6 Azimuth Compression 110
3.6.1 Processing in Azimuth 110
3.6.2 Azimuth Resolution 116
3.6.3 Relation to ISAR 117
3.7 SAR Imaging 118
3.8 SAR Focusing Algorithms 118
3.8.1 RDA 119
3.8.1.1 Range Compression in RDA 120
3.8.1.2 Azimuth Fourier Transform 126
3.8.1.3 Range Cell Migration Correction 128
3.8.1.4 Azimuth Compression 129
3.8.1.5 Simulated SAR Imaging Example 130
3.8.1.6 Drawbacks of RDA 133
3.8.2 Chirp Scaling Algorithm 133
3.8.3 The ω-kA 133
3.8.4 Back-Projection Algorithm 134
3.9 Example of a Real SAR Imagery 135
3.10 Problems in SAR Imaging 136
3.10.1 Range Migration and Range Walk 136
3.10.2 Motion Errors 137
3.10.3 Speckle Noise 140
x Contents
3.11 Advanced Topics in SAR 140

3.11.1 SAR Interferometry 140
3.11.2 SAR Polarimetry 142
References 158
4 Inverse Synthetic Aperture Radar Imaging and Its Basic Concepts 162
4.1 SAR versus ISAR 162
4.2 The Relation of Scattered Field to the Image Function in ISAR 166
4.3 One-Dimensional (1D) Range Profile 167
4.4 1D Cross-Range Profile 172
4.5 Two-Dimensional (2D) ISAR Image Formation (Small Bandwidth, Small
Angle) 176
4.5.1 Resolutions in ISAR 180
4.5.1.1 Range Resolution 181
4.5.1.2 Cross-Range Resolution: 181
4.5.2 Range and Cross-Range Extends 181
4.5.3 Imaging Multibounces in ISAR 182
4.5.4 Sample Design Procedure for ISAR 185
4.5.4.1 ISAR Design Example #1: “Aircraft Target” 189
4.5.4.2 ISAR Design Example #2: “Military Tank Target” 193
4.6 2D ISAR Image Formation (Wide Bandwidth, Large Angles) 197
4.6.1 Direct Integration 198
4.6.2 Polar Reformatting 201
4.7 3D ISAR Image Formation 205
4.7.1 Range and Cross-Range resolutions 209
4.7.2 A Design Example for 3D ISAR 210
References 243
5 Imaging Issues in Inverse Synthetic Aperture Radar 246

5.1 Fourier-Related Issues 246
5.1.1 DFT Revisited 246
5.1.2 Positive and Negative Frequencies in DFT 250
5.2 Image Aliasing 252
5.3 Polar Reformatting Revisited 255
5.3.1 Nearest Neighbor Interpolation 255
5.3.2 Bilinear Interpolation 258
Contents xi
5.4 Zero Padding 260

5.5 Point Spread Function 264
5.6 Windowing 269
5.6.1 Common Windowing Functions 269
5.6.1.1 Rectangular Window 269
5.6.1.2 Triangular Window 269
5.6.1.3 Hanning Window 272
5.6.1.4 Hamming Window 272
5.6.1.5 Kaiser Window 272
5.6.1.6 Blackman Window 276
5.6.1.7 Chebyshev Window 277
5.6.2 ISAR Image Smoothing via Windowing 277
References 304
6 Range-Doppler Inverse Synthetic Aperture Radar Processing 306

6.1 Scenarios for ISAR 306
6.1.1 Imaging Aerial Targets via Ground-Based Radar 307
6.1.2 Imaging Ground/Sea Targets via Aerial Radar 309
6.2 ISAR Waveforms for Range-Doppler Processing 312
6.2.1 Chirp Pulse Train 312
6.2.2 Stepped Frequency Pulse Train 314
6.3 Doppler Shift’s Relation to Cross-Range 316
6.3.1 Doppler Frequency Shift Resolution 317
6.3.2 Resolving Doppler Shift and Cross-Range 318
6.4 Forming the Range-Doppler Image 319
6.5 ISAR Receiver 320
6.5.1 ISAR Receiver for Chirp Pulse Radar 320
6.5.2 ISAR Receiver for SFCW Radar 321
6.6 Quadrature Detection 323
6.6.1 I-Channel Processing 324
6.6.2 Q-Channel Processing 324
6.7 Range Alignment 326
6.8 Defining the Range-Doppler ISAR Imaging Parameters 327
6.8.1 Image Frame Dimension (Image Extends) 327
6.8.2 Range and Cross-Range Resolution 328
6.8.3 Frequency Bandwidth and the Center Frequency 328
6.8.4 Doppler Frequency Bandwidth 328
xii Contents
6.8.5 Pulse Repetition Frequency 329

6.8.6 Coherent Integration (Dwell) Time 329
6.8.7 Pulse Width 330
6.9 Example of Chirp Pulse-Based Range-Doppler ISAR Imaging 331
6.10 Example of SFCW-Based Range-Doppler ISAR Imaging 336
References 347
7 Scattering Center Representation of Inverse Synthetic Aperture

Radar 349
7.1 Scattering/Radiation Center Model 350
7.2 Extraction of Scattering Centers 352
7.2.1 Image Domain Formulation 352
7.2.1.1 Extraction in the Image Domain: The “CLEAN” Algorithm 352
7.2.1.2 Reconstruction in the Image Domain 355
7.2.2 Fourier Domain Formulation 362
7.2.2.1 Extraction in the Fourier Domain 362
7.2.2.2 Reconstruction in the Fourier Domain 364
References 382
8 Motion Compensation for Inverse Synthetic Aperture Radar 385

8.1 Doppler Effect Due to Target Motion 386
8.2 Standard MOCOMP Procedures 388
8.2.1 Translational MOCOMP 389
8.2.1.1 Range Tracking 389
8.2.1.2 Doppler Tracking 390
8.2.2 Rotational MOCOMP 390
8.3 Popular ISAR MOCOMP Techniques 392
8.3.1 Cross-Correlation Method 392
8.3.1.1 Example for the Cross-Correlation Method 394
8.3.2 Minimum Entropy Method 398
8.3.2.1 Definition of Entropy in ISAR Images 398
8.3.2.2 Example for the Minimum Entropy Method 399
8.3.3 JTF-Based MOCOMP 402
8.3.3.1 Received Signal from a Moving Target 403
8.3.3.2 An Algorithm for JTF-Based Rotational MOCOMP 404
8.3.3.3 Example for JTF-Based Rotational MOCOMP 406
Contents xiii
8.3.4 Algorithm for JTF-Based Translational and Rotational MOCOMP 408

8.3.4.1 A Numerical Example 410
References 436
9 Bistatic ISAR Imaging 440

9.1 Why Bi-ISAR Imaging? 440
9.2 Geometry for Bi-Isar Imaging and the Algorithm 444
9.2.1 Bi-ISAR Imaging Algorithm for a Point Scatterer 444
9.2.2 Bistatic ISAR Imaging Algorithm for a Target 448
9.3 Resolutions in Bistatic ISAR 449
9.3.1 Range Resolution 449
9.3.2 Cross-Range Resolution 450
9.3.3 Range and Cross-Range Extends 451
9.4 Design Procedure for Bi-ISAR Imaging 452
9.5 Bi-Isar Imaging Examples 455
9.5.1 Bi-ISAR Design Example #1 455
9.5.2 Bi-ISAR Design Example #2 457
9.6 Mu-ISAR Imaging 465
9.6.1 Challenges in Mu-ISAR Imaging 467
9.6.2 Mu-ISAR Imaging Example 468
References 483
10 Polarimetric ISAR Imaging 484

10.1 Polarization of an Electromagnetic Wave 484
10.1.1 Polarization Type 485
10.1.2 Polarization Sensitivity 486
10.1.3 Polarization in Radar Systems 487
10.2 Polarization Scattering Matrix 488
10.2.1 Relation to RCS 490
10.2.2 Polarization Characteristics of the Scattered Wave 491
10.2.3 Polarimetric Decompositions of EM Wave Scattering 493
10.2.4 The Pauli Decomposition 494
10.2.4.1 Description of Pauli Decomposition 494
10.2.4.2 Interpretation of Pauli Decomposition 495
10.2.4.3 Polarimetric Image Representation Using Pauli Decomposition 496
10.3 Why Polarimetric ISAR Imaging? 497
xiv Contents
10.4 ISAR Imaging with Full Polarization 497

10.4.1 ISAR Data in LP Basis 497
10.4.2 ISAR Data in CP Basis 498
10.5 Polarimetric ISAR Images 499
10.5.1 Pol-ISAR Image of a Benchmark Target 499
10.5.1.1 The “SLICY” Target 499
10.5.1.2 Fully Polarimetric EM Simulation of SLICY 499
10.5.1.3 LP Pol-ISAR Images of SLICY 500
10.5.1.4 CP Pol-ISAR Images of SLICY 502
10.5.1.5 Pauli Decomposition Image of SLICY 503
10.5.2 Pol-ISAR Image of a Complex Target 507
10.5.2.1 The “Military Tank” Target 507
10.5.2.2 Fully Polarimetric EM Simulation of “Tank” Target 508
10.5.2.3 LP Pol-ISAR Images of “Tank” Target 508
10.5.2.4 CP Pol-ISAR Images of “Tank” Target 510
10.5.2.5 Pauli Decomposition Image of “Tank” Target 512
10.6 Feature Extraction from Polarimetric Images 515
References 529
11 Near-Field ISAR Imaging 533

11.1 Definitions of Far and Near-Field Regions 534
11.1.1 The Far-Field Region 534
11.1.1.1 The Far-Field Definition Based on Target’s Cross-Range Extend 534
11.1.1.2 The Far-Field Definition Based on Target’s Range Extend 535
11.1.2 The Near-Field Region 537
11.2 Near-Field Signal Model for the Back-Scattered Field 537
11.3 Near-Field ISAR Imaging Algorithms 540
11.3.1 “Focusing Operator” Algorithm 540
11.3.2 Back-Projection Algorithm 541
11.3.2.1 Fourier Slice Theorem 542
11.3.2.2 BPA Formulation (3D Case) 543
11.3.2.3 BPA Formulation (2D Case) 544
11.4 Data Sampling Criteria and the Resolutions 546
11.5 Near-Field ISAR Imaging Examples 547
11.5.1 Point Scatterers in the Near-Field: Comparison of Far- and Near-Field
Imaging Algorithms 547
11.5.2 Near-Field ISAR Imaging of a Large Object 552
Contents xv
11.5.3 Near-Field ISAR Imaging of a Small Object 555

References 569
12 Some Imaging Applications Based on SAR/ISAR 571

12.1 Imaging Subsurface Objects: GPR-SAR 572
12.1.1 The GPR Problem 572
12.1.2 B-Scan GPR in Comparison to Strip-Map SAR 577
12.1.3 Focused GPR Images Using SAR 577
12.1.3.1 GPR Focusing with ω-k Algorithm (ω-kA) 579
12.1.3.2 GPR Focusing with BPA 582
12.1.3.3 Other Popular GPR Focusing Techniques 589
12.2 Thru-the-Wall Imaging Radar Using SAR 590
12.2.1 Challenges in TWIR 591
12.2.2 Techniques to Improve Cross-Range Resolution in TWIR 591
12.2.3 The Use of SAR in TWIR 592
12.2.4 Example of SAR-Based TWIR 594
12.3 Imaging Antenna-Platform Scattering: ASAR 596
12.3.1 The ASAR Imaging Algorithm 597
12.3.2 Numerical Example for ASAR Imagery 603
12.4 Imaging Platform Coupling Between Antennas: ACSAR 605
12.4.1 The ACSAR Imaging Algorithm 606
12.4.2 Numerical Example for ACSAR 609
12.4.3 Applying ACSAR Concept to the GPR Problem 611
References 615
Appendix 619
Index 628
xvi
Preface to the Second Edition
In the first edition of the book, I tried to cover most of the aspects of inverse syn-
thetic aperture radar (ISAR) imaging starting from Fourier analysis to some
advanced ISAR concepts such as range-Doppler ISAR processing and ISAR motion
compensation techniques. The main goal was to present a conceptual description
of ISAR imagery and the explanation of basic ISAR research topics. Although the
primary audience would be graduate students and other interested researchers in
the fields of electrical engineering and physics, I hoped that colleagues working in
radar research and development or in a related industry might also benefit from
the book.
It has been more than eight years since the publication of the first edition. Since
then, I have been really grateful that I have received positive responses from the
researchers and colleagues that are interested and/or involved in radar imaging,
and especially ISAR imaging. Undoubtedly, ISAR has been gaining more attention
among researchers, scholars, and engineers as emerging new developments in
ISAR research have been reported by various colleagues day by day.
In this second edition of the book, I have tried to include the recent progress
made in ISAR imaging research and also give insights to more advanced concepts.
Therefore, in this edition of the book, I have made the following alterations and
additions:
• All the chapters in the first edition have been revised including all the texts,
equations, and figures with some additions. Typos in the first edition have also
been corrected.
• Chapter 3 that is devoted to the issues of synthetic aperture radar (SAR) has been
extended to include the SAR focusing/processing algorithms such as range-
Doppler algorithm (RDA), back-projection algorithm (BPA) and frequency-
wave number algorithm (ω-kA). The Matlab codes for these algorithms are
being provided with the associated numerical examples. Brief explanations of
Preface to the Second Edition xvii
other SAR focusing/processing algorithms including chirp scaling algorithm

(CSA) and phase shift algorithm (PSA) have also been mentioned.
• In Chapters 4, 5, 7, and 12 where we handle various aspects of ISAR imaging

technology, new scattered field raw data and the corresponding ISAR images
that are more visually attractive are presented together with the associated
Matlab codes that can be used to generate these images.
• A total of three new chapters have been written to cover the topics that were not
considered in the first edition and also to include more detailed subjects of ISAR
imaging to be able to reflect the recent research studies. These are listed below:
– The “Bistatic ISAR (Bi-ISAR) Imaging” concept is covered in Chapter 9. While
the ISAR imaging algorithms presented in previous chapters are based on
monostatic usage of ISAR imaging, we introduce the formulation of ISAR
imaging for the bistatic usages by presenting key aspects such as resolutions
in range/cross-range directions and usage limitations. Also, extension of Bi-
ISAR to multistatic ISAR (Mu-ISAR) imaging is derived with the associated
Matlab examples. A general assessment of Bi-ISAR and Mu-ISAR imaging
to conventional monostatic ISAR imaging is being made throughout the chap-
ter by comparing the outcomes of quantitative metrics and giving the conclud-
ing statements about their advantages and disadvantages based on these
measurable evaluations.
– In Chapter 10, we have added a new and exciting research topic of ISAR called
“Polarimetric ISAR Imaging.” As the traditional ISAR imaging algorithms are
based only on a single polarization of the backscattered electric field, we dem-
onstrate in this chapter that very exciting features of the target can be
extracted with the use of other possible polarizations for the reflected wave.
Polarization decomposition techniques are being introduced and Pauli
decomposition scheme is taken as the tool to be applied to the different polar-
ization ISAR images in this book. The formulation and the usage of Pauli
decomposition technique are presented together with its Matlab codes. Var-
ious realistic simulation examples based on linear polarization, circular polar-
ization, and also Pauli decomposition are given together with obtained
polarimetric ISAR images. It has been demonstrated through the examples
that polarized ISAR images definitely increase the recognition and classifica-
tion of targets by providing increased number of extracted target features.
– Thanks to the recent development in the microwave circuit technology and
antenna design, ISAR imaging algorithms have been started to be used in the
near-field region. Therefore, I have added a new part entitled “Near-field
ISAR imaging” as Chapter 11. The near-field ISAR imaging algorithms are
being introduced. Two of them called “Focusing operator” and the back-
projection based focusing algorithms are given by presenting their theoretical
formulation and algorithm steps together with corresponding Matlab codes.
xviii Preface to the Second Edition
Also, numerical and measured examples based on real scenarios are being
shared.
• In Chapter 12 in which some examples based on SAR/ISAR imaging technol-

ogies are provided, I have previously introduced algorithms called antenna SAR
(ASAR) and antenna coupling SAR (ACSAR) as the unique radar imaging algo-
rithms to image antenna mounted on a platform-to-radar receiver interaction
over the target and to image platform coupling over the antennas mounted
on a target, respectively. In this edition of the book, I have added some new
applications such as ground-penetrating radar (GPR) and through-the-wall ima-
ging radar (TWIR) that also make use of SAR/ISAR imaging algorithms. Meas-
ured examples of GPR and TWIR radar images are provided to demonstrate how
SAR/ISAR imaging algorithms can be effectively used in some popular radar
imaging applications.
I hope that, with the new edition of the book “Inverse Synthetic Aperture Radar
Imaging with MATLAB Algorithms,” the reader would benefit more in terms of
abovementioned new ISAR imaging topics and also from the Matlab codes pro-
vided at the end of chapters.
All MATLAB files may be accessed on the following FTP site: ftp://ftp.wiley.
com/public/sci_tech_med/inverse_synthetic.
Mersin, October 2020 Caner Özdemir

xix
Acknowledgments
I would like to address special thanks to the people below for their help and sup-
port during the preparation of this book. First, I am thankful to my wife, Betül and
my three children for their patience and continuous support while writing this
book. I am very grateful to Dr. Hao Ling, Emeritus Professor in Engineering of
the University of Texas at Austin for being a valuable source of knowledge, ideas,
and also inspiration throughout my academic carreer. He has been a great advisor
since I met him, and his guidance on scientific research is priceless to me.
I would like to express my sincere thanks to my former graduate students; Dr.
Şevket Demirci, Dr. Enes Yiğit, Dr. Betül Yılmaz, Dr. Deniz Üstün, Özkan Kırık,
and Dr. Hakan Işıker who have helped carrying out some of the research presented
in this book. I would also like to thank my graduate student Rasheed Khankan for
his help in preparing references.
Last but not least, I would like to convey my special thanks to Dr. Kai Chang for
inviting me to write the first and then second edition of the book. Without his kind
offer, this book project would not have been possible.
Caner Özdemir
xx
Acronyms
1D One-dimensional
2D Two-dimensional
3D Three-dimensional
ACSAR Antenna coupling synthetic aperture radar
ADC Analog-to-digital converter
ANN Artificial neural network
ASAR Antenna synthetic aperture radar
ATC Automatic target classification
ATR Automatic target recognition
Bi-ISAR Bistatic inverse synthetic aperture radar
BPA Back-projection algorithm
CAD Computer aided design
CDF Cumulative density function
CFAR Constant false alarm rate
CO Contrast optimization
Co-pol Co-polarization
CP Circular polarization
Cross-pol Cross-polarization
CSA Chirp scaling algorithm
CW Continuous wave
DCR Dihedral corner reflectors
DFT Discrete Fourier transform
DTV Digital television
EFIE Electric field integral equation
EM Electromagnetic
ESM Exploding source model
FM Frequency modulated
FMCW Frequency modulated continuous wave
Acronyms xxi
FT Fourier transform
GO Geometric optics
GPR Ground-penetrating radar
GPS Global positioning system
GWN Gaussian white noise
H Horizontal
HH Horizontal–horizontal
HSA Hyperbolic summation algorithm
HV Horizontal–vertical
I Inphase
IDFT Inverse discrete Fourier transform
IFT Inverse Fourier transform
IMU Inertial measurement unit
InSAR Interferometric SAR
ISAR Inverse synthetic aperture radar
JTF Joint time-frequency
KB Kbytes
KMA Kirchhoff migration algorithm
L Left
LFM Linear frequency modulated
LFMCW Linear frequency modulated continuous wave
LHCP Left-hand circular polarized
LHEP Left-hand elliptically polarized
LL Left–left
LOS Line of sight
LP Linear polarization
LR Left–right
MB Mbytes
MDA Map-drift autofocus
MFIE Magnetic field integral equation
MIMO Multiple-input multiple-output
MOCOMP Motion compensation
Mu-ISAR Multi-static inverse synthetic aperture radar
PEC Perfect electric conductor
PGA Phase gradient autofocus
P-ISAR Passive inverse synthetic aperture radar
PO Physical optics
PolSAR Polarimetric synthetic aperture radar
PPP Prominent point processing
PRF Pulse repetition frequency
xxii Acronyms
PRI Pulse repetition interval

PSD Power spectral density
PSF Point spread function
PSLR Peak-to-sidelobe ratio
PSMA Phase-shift migration algorithm
PSR Point-spread-response
Q Quadrature
QD Quadradure detection
R Right
RCM Range cell migration
RCMC Range cell migration correction
RCS Radar cross section
RDA Range-doppler algorithm
RF Radiofrequency
RGB Red green blue
RHCP Right-hand circular polarized
RHEP Right-hand elliptically polarized
RL Right–Left
RLOS Radar line of sight
RR Right–right
Rx Receiver
[S] Polarization scattering matrix
SAC Shift and correlate
SAR Synthetic aperture radar
SBR Shooting and bouncing ray
SFCW Stepped frequency continuous wave
sinc Sinus cardinalis
SNR Signal-to-noise ratio
SPU Signal processing unit
STFT Short-time Fourier transform
TCR Trihedral corner reflector
TFDS Time-frequency distribution series
TWIR Through-the-wall imaging radar
TWR Through-the-wall radar
Tx Transmitter
V Vertical
VH Vertical–horizontal
VNA Vector network analyzer
VV Vertical–vertical
ω-kA Frequency-wavenumber algorithm
1
Basics of Fourier Analysis
1.1 Forward and Inverse Fourier Transform
Fourier transform (FT) is a common and useful mathematical tool that is utilized
in innumerous applications in science and technology. FT is quite practical espe-
cially for characterizing nonlinear functions in nonlinear systems, analyzing ran-
dom signals, and solving linear problems. FT is also a very important tool in radar
imaging applications as we shall investigate in the forthcoming chapters of this
book. Before starting to deal with the FT and inverse Fourier transform (IFT), a
brief history of this useful linear operator, and its founders are presented.
1.1.1 Brief History of FT

Jean Baptiste Joseph Fourier, a great mathematician, was born in 1768, Auxerre,
France. His special interest in heat conduction led him to describe a mathematical
series of sine and cosine terms that could be used to analyze propagation and dif-
fusion of heat in solid bodies. In 1807, he tried to share his innovative ideas with
researchers by preparing an essay entitled as On the Propagation of Heat in Solid
Bodies. The work was examined by Lagrange, Laplace, Monge, and Lacroix.
Lagrange’s oppositions caused the rejection of Fourier’s paper. This unfortunate
decision cost colleagues to wait for 15 more years to meet his remarkable contri-
butions to mathematics, physics, and especially on signal analysis. Finally, his
ideas were published thru the book The Analytic Theory of Heat in 1822
(Fourier 1955).
Discrete Fourier transform (DFT) was developed as an effective tool in calculat-
ing this transformation. However, computing FT with this tool in the nineteenth
century was taking a long time. In 1903, C. Runge has studied on the minimization
Inverse Synthetic Aperture Radar Imaging with MATLAB Algorithms: With Advanced SAR/ISAR
Imaging Concepts, Algorithms, and MATLAB Codes, Second Edition. Caner Özdemir.
© 2021 John Wiley & Sons, Inc. Published 2021 by John Wiley & Sons, Inc.
2 1 Basics of Fourier Analysis
of the computational time of the transformation operation (Runge 1903). In 1942,

Danielson and Lanczos had utilized the symmetry properties of FT to reduce the
number of operations in DFT (Danielson and Lanczos 1942). Before the advent of
digital computing technologies, James W. Cooley and John W. Tukey developed a
fast method to reduce the computation time of DFT operation. In 1965, they
published their technique that later on has become famous as the fast Fourier
transform (FFT) (Cooley and Tukey 1965).
1.1.2 Forward FT Operation

The FT can be simply defined as a certain linear operator that maps functions or
signals defined in one domain to other functions or signals in another domain. The
common use of FT in electrical engineering is to transform signals from time
domain to frequency domain or vice-versa. More precisely, forward FT decom-
poses a signal into a continuous spectrum of its frequency components such that
the time signal is transformed to a frequency domain signal. In radar applications,
these two opposing domains are usually represented as “spatial-frequency (or
wave-number)” and “range (distance).” Such use of FT will be often examined
and applied throughout this book.
The forward FT of a continuous signal g(t) where −∞ < t < ∞ is described as
G f = gt
∞ 11
= gt exp − j2πft dt
−∞
where represents the forward FT operation that is defined from time domain to
frequency domain.
To appreciate the meaning of FT, the multiplying function exp(−j2πft) and
operators (multiplication and integration) on the right of side of Eq. 1.1 should
be examined carefully: The term exp − j2π f i t is a complex phasor representation
for a sinusoidal function with the single frequency of “fi.” This signal oscillates
with the single frequency of “fi” and does not contain any other frequency com-
ponent. Multiplying the signal in interest, g(t) with exp − j2π f i t provides the sim-
ilarity between each signal, that is, how much of g(t) has the frequency content of
“fi.” Integrating this multiplication over all time instants from −∞ to ∞ will sum
the “fi” contents of g(t) over all time instants to give G(fi) that is the amplitude of
the signal at the particular frequency of “fi.” Repeating this process for all the fre-
quencies from −∞ to ∞ will provide the frequency spectrum of the signal repre-
sented as G(f). Therefore, the transformed signal represents the continuous
spectrum of frequency components; i.e. representation of the signal in “frequency
domain.”
1.2 FT Rules and Pairs 3
1.1.3 IFT
This transformation is the inverse operation of the FT. IFT, therefore, synthesizes
a frequency-domain signal from its spectrum of frequency components to its
time domain form. The IFT of a continuous signal G(f) where −∞ < f < ∞ is
described as
−1
gt = G f
∞ 12
= G f exp j2πft df
−∞
where the IFT operation from frequency domain to time domain is represented by
−1
.
1.2 FT Rules and Pairs
There are many useful Fourier rules and pairs that can be very helpful when apply-
ing the FT or IFT to different real-world applications. We will briefly revisit them
to remind the properties of the FT to the reader. Provided that FT and IFT are
defined as in Eqs. 1.1 and 1.2, respectively, FT pair is denoted as
gt G f 13
and the corresponding alternative pair is given by
−1
G f gt 14
Based on these notations, the properties of FT are listed briefly below.
1.2.1 Linearity
If G(f) and H(f) are the FTs of the time signals g(t) and h(t), respectively, the fol-
lowing equation is valid for the scalars a and b.
a g t +b h t a G f +b H f 15
Therefore, the FT is a linear operator.
1.2.2 Time Shifting

If the signal is shifted in time with a value of to, then the corresponding frequency
signal will have the form of
g t − to exp − j2πf t o G f 16
1.2.3 Frequency Shifting

If the time signal is multiplied by a phase term of exp j2π f o t , then the FT of this
time signal is shifted in frequency by fo as given below
exp j2π f o t gt G f − fo 17
1.2.4 Scaling
If the time signal is scaled by a constant a, then the spectrum is also scaled with the
following rule
1 f
g at G , a R, a 0 18
a a
1.2.5 Duality
If the spectrum signal G(f) is taken as a time signal G(t), then, the corresponding
frequency domain signal will be the time reversal equivalent of the original time
domain signal, g(t) as
Gt g −f 19
1.2.6 Time Reversal

If the time is reversed for the time-domain signal, then the frequency is also
reversed in the frequency domain signal.
g −t G −f 1 10
1.2.7 Conjugation
If the conjugate of the time-domain signal is taken, then the frequency-domain
signal conjugated and frequency-reversed.
g∗ t G∗ − f 1 11
1.2.8 Multiplication
If the time-domain signals, g(t) and h(t) are multiplied in time, then their spectrum
signals G(f) and H(f) are convolved in frequency.
gt ht G f ∗H f 1 12
1.2.9 Convolution
If the time-domain signals, g(t) and h(t) are convolved in time, then their spectrum
signals G(f) and H(f) are multiplied in the frequency domain.
g t ∗h t G f H f 1 13
1.2.10 Modulation
If the time-domain signal is modulated with sinusoidal functions, then the fre-
quency-domain signal is shifted by the amount of the frequency at that particular
sinusoidal function.
1
gt cos 2π f o t G f + fo + G f − fo
2
1 14
j
gt sin 2π f o t G f + fo −G f − fo
2
1.2.11 Derivation and Integration

If the derivative or integration of a time-domain signal is taken, then the corre-
sponding frequency-domain signal is given as below.
d
gt 2πf G f
dt
t 1 15
1
g τ dτ G f + πG 0 δ f
−∞ j2πf
1.2.12 Parseval’s Relationship

A useful property that was claimed by Parseval is that since the FT (or IFT) oper-
ation maps a signal in one domain to another domain, their energies should be
exactly the same as given by the following relationship.
∞ ∞
2 2
gt dt G f df 1 16
−∞ −∞
1.3 Time-Frequency Representation of a Signal
While the FT concept can be successfully utilized for the stationary signals, there
are many real-world signals whose frequency contents vary over time. To be able to
display these frequency variations over time; therefore, joint time–frequency (JTF)
transforms/representations are being used.
1.3.1 Signal in the Time Domain

The term “time domain” is used while describing functions or physical signals
with respect to time either continuous or discrete. The time-domain signals are
usually more comprehensible than the frequency-domain signals since most of
the real-world signals are recorded and displayed versus time. Common equip-
ment is to analyze time-domain signals is the oscilloscope. In Figure 1.1, a time-
domain sound signal is shown. This signal is obtained by recording of an utterance
of the word “prince” by a lady. By looking at the occurrence instants in the x-axis
and the signal magnitude in the y-axis, one can analyze the stress of the letters
inside the word “prince.”
1.3.2 Signal in the Frequency Domain

The term “frequency domain” is used while describing functions or physical sig-
nals with respect to frequency either continuous or discrete. Frequency-domain
representation has been proven to be very useful in innumerous engineering appli-
cations while characterizing, interpreting, and identifying signals. Solving differ-
ential equations, analyzing circuits, signal analysis in communication systems are
few among many others where frequency-domain representation is much more
Time domain signal
0.2
0.1
0
Amplitude
–0.1
–0.2
–0.3
–0.4
–0.5
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

Time, s
Figure 1.1 The time-domain signal of “prince” spoken by a lady.

Frequency domain signal

0.035
0.35
0.025
0.25
Amplitude
0.015
0.15
0.01
0.005
0
0 5 10 15 20
Frequency, kHz
Figure 1.2 The frequency-domain signal (or the spectrum) of “prince.”
advantageous than the time-domain representation. The frequency-domain signal

is traditionally obtained by taking the FT of the time-domain signal. As briefly
explained in Section 1.1, FT is generated by expressing the signal onto a set of basis
functions, each of which is a sinusoid with the unique frequency. Displaying the
measure of the similarities of the original time-domain signal to those particular
unique-frequency bases generates the Fourier transformed signal or the fre-
quency-domain signal. Spectrum analyzers and network analyzers are the common
equipments to analyze frequency-domain signals. These signals are not as quite
perceivable when compared to time-domain signals. In Figure 1.2, the fre-
quency-domain version of the sound signal in Figure 1.1 is obtained by using
the DFT operation. The signal intensity value at each frequency component can
be read from the y-axis. The frequency content of a signal is also called the spectrum
of that signal.
1.3.3 Signal in the Joint Time-Frequency (JTF) Plane

Although FT is very effective for demonstrating the frequency content of a sig-
nal, it does not give the knowledge of frequency variation over time. However,
most of the real-world signals have time-varying frequency content such as
speech and music signals. In these cases, the single-frequency sinusoidal bases
are not considered to be suitable for the detailed analysis of those signals.
Therefore, JTF analysis methods were developed to represent these signals both
in time and frequency to observe the variation of frequency content as the time
progresses.
There are many tools to map a time domain or frequency-domain signal onto
the JFT plane. Some of the most well-known JFT tools are short-time Fourier
transform (STFT) (Allen 1977), Wigner–Ville distribution (Nuttall 1988),
Choi–Willams distribution (Du and Su 2003), Cohen’s class (Cohen 1989),
and time-frequency distribution series (TFDS) (Qian and Chen 1996). Among
these, the most appreciated and commonly used one is the STFT or the spectro-
gram. STFT can easily display the variations in the sinusoidal frequency and
phase content of local moments of a signal over time with sufficient resolution
in most cases.
The spectrogram transforms the signal onto two-dimensional (2D) time-
frequency plane via the following famous equation:
STFT g t ≜ G t, f
∞ 1 17
= gτ w τ − t exp − j2πf τ dτ
−∞
This transformation formula is nothing but the short-time (or short-term) ver-
sion of the famous FT operation defined in Eq. 1.1. The main signal, g(t) is multi-
plied with a shorter duration window signal, w(t). By sliding this window signal
over g(t) and taking the FT of the product, only the frequency content for the win-
dowed version of the original signal is acquired. Therefore, after completing the
sliding process over the whole duration of the time-domain signal g(t) and putting
corresponding FTs side by side, the 2D STFT of g(t) is obtained.
It is obvious that STFT will produce different output signals for different dura-
tion windows. The duration of the window affects the resolutions in both domains.
While a very short-duration time window provides a good resolution in the time
direction, the resolution in the frequency direction becomes poor. This is because
of the fact that the time duration and the frequency bandwidth of a signal are
inversely proportional to each other. Similarly, a long duration time signal will
give a good resolution in frequency domain while the resolution in the time
domain will be bad. Therefore, a reasonable selection has to be bargained about
the duration of the window in time to be able to view both domains with fairly
good enough resolutions.
The shape of the window function has an effect on the resolutions as well. If a
window with sharp ends is chosen, there will be strong side lobes in the other
domain. Therefore, smooth waveform type windows are usually utilized to obtain
well-resolved images with less side lobes with the price of increased main
Signal in time-frequency plane

16
14
12
Frequency, kHz
10
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Time, s
Figure 1.3 The time-frequency representation of the word “prince.”
beamwidth, i.e. less resolution. Commonly used window types are Hanning, Ham-
ming, Kaiser, Blackman, and Gaussian.
An example of the use of spectrogram is demonstrated in Figure 1.3. The spec-
trogram of the sound signal in Figure 1.1 is obtained by applying the STFT oper-
ation with a Hanning window. This JFT representation obviously demonstrates
the frequency content of different syllables when the word “prince” is spoken.
Figure 1.3 illustrates that while the frequency content of the part “prin…” takes
place at low frequencies, that of the part “..ce” occurs at much higher frequencies.
JTF transformation tools have been found to be very useful in interpreting the
physical mechanisms such as scattering and resonance for radar applications
(Trintinalia and Ling 1995; Filindras et al. 1996; Özdemir and Ling 1997; Chen
and Ling 2002). In particular, when JTF transforms are used to form the 2D image
of electromagnetic scattering from various structures, many useful physical fea-
tures can be displayed. Distinct time events (such as scattering from point targets
or specular points) show up as vertical line in the JTF plane as depicted in
Figure 1.4a. Therefore, these scattering centers appear at only one time instant
but for all frequencies. A resonance behavior such as scattering from an open cav-
ity structure shows up as horizontal line on the JTF plane. Such mechanisms occur
only at discrete frequencies but over all time instants (see Figure 1.4b). Dispersive
mechanisms, on the other hand, are represented on the JTF plane as slanted
curves. If the dispersion is due to the material, then the slope of the image is
b
Frequency
e
d
f
c
a
Time
Figure 1.4 Images of scattering mechanisms in the joint time–frequency plane. (a)
Scattering center, (b) resonance, (c and d) dispersion due to material, (e and f ) dispersion due
to geometry of the structure.
positive as shown in Figure 1.4c,d. The dielectric coated structures are the good
examples of this type of dispersion. The reason for having a slanted line is because
of the modes excited inside such materials. As frequency increases, the wave veloc-
ity changes for different modes inside these materials. Consequently, these modes
show up as slanted curves in the JTF plane. Finally, if the dispersion is due to the
geometry of the structure, this type of mechanism appears as a slanted line with a
negative slope. This style of behavior occurs for such structures such as waveguides
where there exist different modes with different wave velocities as the frequency
changes as seen in Figure 1.4e,f.
An example of the use of JTF processing in radar application is shown in
Figure 1.5 where spectrogram of the simulated backscattered data from a dielec-
tric-coated wire antenna is shown (Özdemir and Ling 1997). The backscattered
field is collected from the Teflon-coated wire (εr = 2.1) such that the tip of the
electric field makes an angle of 60 with the wire axis as illustrated in Figure 1.5.
After the incident field hits the wire, infinitely successive scattering mechan-
isms occur. The first four of them are illustrated on top of Figure 1.5. The first
return comes from the near tip of the wire. This event occurs at a discrete time
that exists at all frequencies. Therefore, this return demonstrates a scattering
center-type mechanism. On the other hand, all other returns experience at least
one trip along the dielectric-coated wire. Therefore, they confront a dispersive
behavior. As the wave travels along the dielectric-coated wire, it is influenced by
the dominant dispersive surface mode called Goubau (Richmond and Newman
1976). Therefore, the wave velocity decreases as the frequency increases such
that the dispersive returns are tilted to later times on the JTF plane. The dom-
inant dispersive scattering mechanisms numbered 2, 3, and 4 are illustrated in
Figure 1.5 where the spectrogram of the backscattered field is presented. The
other dispersive returns with decreasing energy levels can also be easily
1.4 Convolution and Multiplication Using FT 11
3
30°
1
1 2 3 4
0
16
–10
Frequency, GHz
14
12 –20
10
–30
8
–40
0 5 10 15 dB
Time, ns
Scattering Dispersive
center returns
Figure 1.5 JTF image of a backscattered measured data from a dielectric-coated wire
antenna using spectrogram.
observed from the spectrogram plot. As the wave travels on the dielectric-coated
wire more and more, it is slanted more on the JTF plane, as expected.
1.4 Convolution and Multiplication Using FT
Convolution and multiplication of signals are often used in radar signal proces-
sing. As listed in Eqs. 1.12 and 1.13, convolution is the inverse operation of mul-
tiplication as the FT is concerned, and vice versa. This useful feature of the FT is
widely used in signal and image processing applications. It is obvious that the mul-
tiplication operation is significantly faster and easier to deal with when compared
to the convolution operation, especially for long signals. Instead of directly convol-
ving two signals in the time domain, therefore, it is much easier and faster to take
the IFT of the multiplication of the spectrums of those signals as shown below:
−1
g t ∗h t = gt ht
1 18
−1
= G f H f
In a dual manner, convolution between the frequency-domain signals can be

calculated in a much faster and easier way by taking the FT of the product of their
time-domain versions as formulated below:
−1 −1
G f ∗H f = G f H f
1 19
= gt ht
1.5 Filtering/Windowing
Filtering is the common procedure that is used to remove undesired parts of sig-
nals such as noise. It is also used to extract some useful features of the signals. The
filtering function is usually in the form of a window in the frequency domain.
Depending on the frequency inclusion of the window in the frequency axis, the
filters are named low-pass (LP), high-pass (HP), or band-pass (BP).
The frequency characteristics of an ideal LP filter are depicted as dashed line in
Figure 1.6. Ideally, this filter should pass frequencies from DC to the cut-off fre-
quency; fc and should stop higher frequencies beyond. In real practice, however,
ideal LP filter characteristics cannot be realized. According to the Fourier theory, a
signal cannot be both time limited and band limited. That is to say, to be able to
Filter
amplitude
Ideal
0 dB
Real
–3 dB
Pass-band Stop-band
fR fC Frequency
Figure 1.6 An ideal and real LP filter characteristics.

1.5 Filtering/ Windowing 13
achieve an ideal band-limited characteristic as in Figure 1.6, then the correspond-

ing time-domain signal should theoretically extent to infinity which is of course
not possible for realistic applications. Since all practical human-made signals
are time limited, i.e. it should start and stop at specific time instants, the frequency
contents of these signals extent to infinity. Therefore, an ideal filter characteristic
as the one in Figure 1.6 cannot be realizable; but, only the approximate versions of
it can be implemented in real applications. The best implementation of practical
low-pass filter characteristic was achieved by Butterworth (Daniels 1974) and Che-
byshev (Williams and Taylors 1988). The solid line in Figure 1.6 demonstrates a
real LP filter characteristic of Butterworth type.
Windowing procedure is usually applied to smoothen a time-domain signal,
therefore, filtering out higher frequency components. Some of the popular win-
dows that are widely used in signal and image processing are Kaiser, Hanning,
Hamming, Blackman, and Gaussian. A comparative plot of some of these windows
is given in Figure 1.7.
The effect of windowing operation is illustrated in Figure 1.8. A time-domain
signal of a rectangular signal is shown in Figure 1.8a and its FT is provided in
Figure 1.8b. This function is, in fact, a sinc (sinus cardinalis) function and has
major side lobes. For the sinc function, the highest side lobe is approximately
13 dB lower than the apex of the main lobe. This much of contrast, of course,
may not be sufficient in some imaging applications. As shown in Figure 1.8c,
the original rectangular time-domain signal is Hanning windowed. Its
0.8
0.6
Amplitude
Kaiser
0.4
Hanning
Hamming
0.2
0
0 20 40 60 80 100
Frequency
Figure 1.7 Some common window characteristics.

(a) (b)
1
0.3
0.8
0.6 0.2
0.4
0.1
0.2
0 0
–50 0 50 –40 –20 0 20 40
Time, s Frequency, Hz
(c) (d)
1
0.3
0.8
0.6 0.2
0.4
0.1
0.2
0 0
–50 0 50 –40 –20 0 20 40
Time, s Frequency, Hz
Figure 1.8 Effect of windowing. (a) Rectangular time signal, (b) its Fourier spectrum: a sinc
signal, (c) Hanning windowed time signal, (d) corresponding frequency-domain signal.
corresponding spectrum is depicted in Figure 1.8d where the side lobes are highly
suppressed thanks to the windowing operation. For this example, the highest side
lobe level is now 32 dB below the maximum value of the main lobe which pro-
vides better contrast when compared to the original, non-windowed signal.
A main drawback of windowing is the resolution decline in the frequency signal.
The FT of the windowed signal has worse resolution than the FT of the original
time-domain signal. This feature can also be noticed from the example in
Figure 1.8. By comparing the main lobes of the figures on the right, the resolution
after windowing is almost twice as bad when compared to the original frequency-
domain signal. A comprehensive examination of windowing procedure will be
presented later on, in Chapter 5.
1.6 Data Sampling
Sampling can be regarded as the preprocess of transforming a continuous or ana-

log signal to a discrete or digital signal. When the signal analysis has to be done
using digital computers via numerical evaluations, continuous signals need to be
1.6 Data Sampling 15
Figure 1.9 Sampling. (a) (a)

continuous time signal, (b) s(t)
discrete-time signal after the
sampling.
(b)
s[n]
Sampling
instants
0Ts Ts 2Ts 3Ts 4Ts . . . . . t
0 1 2 3 4 ... . . n
Sampling
numbers
converted to the digital versions. This is achieved by applying the common proce-
dure of sampling. Analog-to-digital (A/D) converters are common electronic
devices to accomplish this process. The implementation of a typical sampling proc-
ess is shown in Figure 1.9. A time signal s(t) is sampled at every Ts seconds such
that the discrete signal, s[n], is generated via the following equation:
s n = s nT s , n = 0, 1, 2, 3, … 1 20
Therefore, the sampling frequency fs is equal to 1/Ts where Ts is called the sam-
pling interval.
A sampled signal can also be regarded as the digitized version of the multipli-
cation of the continuous signal, s(t) with the impulse comb waveform, c(t) as
depicted in Figure 1.10.
According to the Nyquist–Shannon sampling theorem, the perfect reconstruc-
tion of the signal is only possible provided that the sampling frequency fs is equal
or larger than twice the maximum frequency content of the sampled signal
Figure 1.10 Impulse comb waveform c(t)

composed of ideal impulses.
... ...
0 Ts 2Ts 3Ts 4Ts . . . . .

(Shannon 1949). Otherwise, signal aliasing is unavoidable and only distorted

version of the original signal can be reconstructed.
1.7 DFT and FFT
1.7.1 DFT
As explained in Section 1.1, the FT is used to transform continuous signals from
one domain to another. It is usually used to describe the continuous spectrum of an
aperiodic time signal. To be able to utilize the FT while working with digital sig-
nals, the digital or DFT has to be used.
Let s(t) be a continuous periodic time signal with a period of To = 1/fo. Then, its
sampled (or discrete) version is s[n] ≜ s(nTs) with a period of NTs = To where N is
the number of samples in one period. Then, the Fourier integral in Eq. 1.1 will turn
to a summation as shown below.
N −1
S kfo = s nT s exp − j2π k f o nT s
n=0
N −1
k
= s nT s exp − j2π nT s 1 21
n=0
NT s
N −1
k
= s nT s exp − j2π n
n=0
N
Dropping the fo and Ts inside the parenthesis for the simplicity of nomenclature
and therefore switching to discrete notation, DFT of the discrete signal s[n] can be
written as
N −1
k
Sk = s n exp − j2π n 1 22
n=0
N
In a dual manner, let S(f) represent a continuous periodic frequency signal with
a period of Nfo = N/To and let S[k] ≜ S(kfo) be the sampled signal with the period of
Nfo = fs. Then, the IDFT of the frequency signal S[k] is given by
N −1
n n
s = S kfo exp j2π k f o
fs k=0
fs
N −1
n
s nT s = S kfo exp j2π k f o 1 23
k=0
Nf o
N −1
n
= S kfo exp j2π k
k=0
N
1.7 DFT and FFT 17
Using the discrete notation by dropping the fo and Ts inside the parenthesis, the
IDFT of a discrete frequency signal S[k] is given as
N −1
n
sn = Sk exp j2π k 1 24
k=0
N
1.7.2 FFT
FFT is the efficient and fast way of evaluating the DFT of a signal. Normally, com-
puting the DFT is in the order of N2 arithmetic operations. On the other hand, fast
algorithms like Cooley-Tukey’s FFT technique produce arithmetic operations in
the order of N log(N) (Cooley and Tukey 1965; Brenner and Rader 1976; Duhamel
1990). An example of DFT is given is Figure 1.11 where a discrete time-domain
ramp signal is plotted in Figure 1.11a and its frequency-domain signal obtained
by an FFT algorithm is given in Figure 1.11b.
1.7.3 Bandwidth and Resolutions

The duration, the bandwidth, and the resolution are important parameters while
transforming signals from time domain to frequency domain or vice versa. Con-
sidering a discrete time-domain signal with a duration of To = 1/fo sampled N times
with a sampling interval of Ts = To/N, the frequency resolution (or the sampling
interval in frequency) after applying the DFT can be found as
1
Δf = 1 25
To
The spectral extend (or the frequency bandwidth) of the discrete frequency sig-
nal is
B = N Δf
N
= 1 26
To
1
=
Ts
For the example in Figure 1.11, the signal duration is 1 ms with N = 10 samples.
Therefore, the sampling interval is 0.1 ms. After applying the expressions in
Eqs. 1.25 and 1.26, the frequency resolution is 100 Hz and the frequency band-
width is 1000 Hz. After taking the DFT of the discrete time-domain signal, the first
entry of the discrete frequency signal corresponds to zero frequency and negative
frequencies are located in the second half of the discrete frequency signal as seen in
Figure 1.11b. After the DFT operation, therefore, the entries of the discrete
(a) 1.2 Figure 1.11 An example of

DFT operation: (a) discrete time-
1 domain signal, (b) discrete
frequency-domain signal
0.8 without FFT shifting, (c) discrete
frequency-domain signal with
FFT shifting.
s[n]
0.6
0.4
0.2
0
0 2 4 6 8 10
Time, ms
(b)
6
4
s[k]
0
0 200 400 600 800 1000
Frequency, Hz
(c)
6
4
s[k]
0
–500 0 500
Frequency, Hz
frequency signal should be swapped from the middle to be able to form the fre-
quency axis correctly as shown in Figure 1.11c. This property of DFT will be thor-
oughly explored in Chapter 5 to demonstrate its use in ISAR imaging.
Similar arguments can be made for the case of IDFT. Considering a discrete fre-
quency-domain signal with a bandwidth of B sampled N times with a sampling
interval of Δf, the time resolution (or the sampling interval in time) after applying
IDFT can be found as
Δt = T s
1
= 1 27
B
1
=
NΔf
The time duration of the discrete time signal is
1
To = 1 28
Δf
For the frequency-domain signal in Figure 1.11b or c, the frequency bandwidth
is 1000 Hz with N = 10 samples. Therefore, the sampling interval in frequency is
100 Hz. After applying IDFT to get the time-domain signal as in Figure 1.11a, the
formulas in Eqs. 1.27 and 1.28 calculate the resolution in time as 0.1 ms and the
duration of the signal as 1 ms.
1.8 Aliasing
Aliasing is a type of signal distortion due to undersampling the signal of interest.

According to Nyquist–Shannon sampling theorem (Shannon 1949), the sampling
frequency fs should be equal to or larger than twice the maximum frequency con-
tent; fmax of the signal to be able to perfectly reconstruct the original signal:
f s ≥ 2 f max 1 29
Since processing the analog radar signal requires sampling of the received data,
the concept of aliasing should be taken into account when dealing with radar
signals.
1.9 Importance of FT in Radar Imaging
The imaging of a target using electromagnetic waves emitted from radars is mainly
based on the phase information of the scattered waves from the target. This is
because of the fact that the phase information is directly related to the range dis-
tance of the target. In the case of monostatic radar configuration as shown in
Figure 1.12a, let the scattering center on the target be at R distance away from
the radar. Then, the scattered field E s from this scattering center on the target
has a complex scattering amplitude, A and a phase factor that contains the distance
information of the target as follows:
Es A exp − j2kR 1 30
As is obvious from Eq. 1.30, there exists a Fourier relationship between the wave
number, k, and the distance, R. Provided that the scattered field is collected over a
bandwidth of frequencies (Figure 1.12b), it is possible to pinpoint the distance R by
(a) Figure 1.12 (a) Monostatic

radar configuration, (b)
scattered field versus
R
Rmin Target frequency, (c) range profile
Rmax of the target.
Radar
(b)
Es
fmin fmax f
(c)
Range
profile
Rmin Rmax
Fourier transforming the scattered field data as depicted in Figure 1.12c. The plot
of scattered field versus range is called the range profile, which is an important phe-
nomenon in radar imaging. Range profile is, in fact, nothing but the one-
dimensional range image of the target. An example is illustrated in Figure 1.13
Figure 1.13 Simulated range

profile of an airplane.
× 10–3
1
0.8
0.6
0.4
0.2
0
0 2 4 6 8 10 12
Range, m
Another random document with
no related content on Scribd:
were built there in his time, and during his reign and for many a long year
after it was the custom to hire merchant vessels and arm them—if they
were not already armed to protect themselves against pirates—to
augment the national fleet. Religious, no less than national, rivalry
contributed, albeit unconsciously, to the development of the efficacy of the
warship as a fighting unit. The enmity between Britain and Spain
culminated, in Elizabeth’s reign, after a series of daring attacks by
reckless Englishmen upon the Spanish fleet in preparation for the great
attack upon England, in the dispatch of the Armada. Hawkins and one or
two others foresaw that the advantage would lie with the fleet which could
be most effectively manœuvred. The disparity between the Armada and
the British fleet was not so great as many writers have represented, either
in the size or number of the vessels; but the British vessels on the
average were smaller, faster, and better handled; in other words,
efficiency told against sheer weight of numbers. This was the last great
sea-fight on the ocean in which oared ships took part; they were no match
for their smaller and more speedy sailing antagonists.
Structurally, most of the vessels of this time, the larger especially, were
disfigured by high sterncastles, but early in the seventeenth century this
encumbrance and many others had disappeared. Thence to the
nineteenth century the development of warships was marked mainly by
continual increases in their size, improving their form of hull and,
consequently, their speed and buoyancy; augmenting their sail area and
perfecting the square-rigged system; and adding to the number of gun
decks and the number of guns carried; until the grand wooden three-
deckers swept the seas in all their ponderous pride and majesty. Ships of
the line of various ratings played their part, and were ably seconded by
frigates, brigs, cutters, sloops and bomb-ketches. All these were in vogue
less than a century ago, and though not forgotten, are looked upon as
historical and romantic and interesting curiosities.
* * * * * *
In the weapons, no less than in the ships, the changes have been
marvellous. For many centuries after ships were adopted for war, the
fighting was done by soldiers carried aboard them. The human machines,
the rowers, had to attend exclusively to their oars, for on them the safety
or success of the fighting men depended. The main idea was to get to
close quarters and fight hand to hand with javelin or sword, spear or
battleaxe; bows and arrows were used when possible, and missiles
hurled by hand were not despised. The ram, in various forms, affixed to
the bows in such a manner as to strike the enemy’s ship below or above
the water-line, or both, was used with fearful effect in many a stubbornly
fought engagement.
The introduction of artillery in the fourteenth century marked the beginning
of the first great revolution in naval warfare, and the changes in the
projectiles have been no less extraordinary than those in the guns.
* * * * * *
The next great revolution was the introduction of the steam-engine. Its
adoption in the British Navy in 1832 marked the beginning of the end of
the sailing warship. Her last grim battle against inexorable fate was fought
with the same doggedness which had distinguished her in many an
encounter with her nation’s enemies; but the superiority of steam over sail
was recognised. Temporising measures, a patched-up peace, as it were,
lasted for a few years while the steam engine was employed as an
auxiliary. Sail power, however, had reached its apotheosis so far as
warships were concerned. Engineers, animated by practical common
sense and ignoring romantic associations, improved their engines, so that
the steam power was no longer the assistant of sail, but its associate, and
was quick in attaining the position of chief partner and showing that sails
could be dispensed with altogether. The Crimean War sounded the knell
of the wooden battleship as well as of the paddle-wheel war steamer. The
former gave place to the iron-clad vessel, and the latter was supplanted
by the screw-propelled ship. The power of artillery had shared in the
application of scientific knowledge and benefited accordingly. The great
battle between the maker of armour plates and the maker of guns and
projectiles had begun. Iron, the conqueror of wood, had but a short reign.
Where iron was used a few years ago, steel is now invariably employed.
The thoroughness of the victory is shown by the fact that in the whole of
the British Isles not one iron vessel, large or small, was built in 1909 for
war or commerce.
* * * * * *
The years 1905 and 1906 saw two of the most important steps forward in
the history of warships, for they included the adoption of the turbine
principle of warship propulsion and the “Dreadnought” principle of
armament. The progress of the last fifty years, culminating in the
Dreadnoughts, has been wonderful; already designs of vessels intended
to relegate them to second place are under consideration.
Type after type of battleship, cruiser, scout, gunboat, destroyer and
torpedo-boat has followed in rapid succession of late years, and
submarines have become an accomplished fact. He would be a foolish
man who would prophesy that the end is in sight.
There is nothing more marvellous in the world’s history than the
tremendous development in marine engineering, in warship construction,
in explosives, in armament, and in projectiles that has taken place in the
latter half of the nineteenth century, and especially in the last twenty-five
years.
WARSHIPS AND THEIR STORY
CHAPTER I
FROM ANCIENT EGYPT TO THE INTRODUCTION OF ARTILLERY
When did man first entrust himself afloat for purposes of war? and
what was the type of vessel he employed? are questions which take
us back almost to the earliest stages of historical human progress,
concerning which all the knowledge of the antiquaries is but
conjectural, a stage so remote that scientists have not yet
determined how many thousands of years ago it existed. The earliest
vessels thus employed must have been transports, and nothing else;
but if employed as aids to aggression when the kings of the earth
took counsel together and, impelled by avarice or a desire to assist
in one another’s turbulent love affairs, or, for their own safety,
convinced of the necessity of finding an outlet for the energies of
their restless subjects, invaded the territories of their neighbours, the
ships, whatever their nature, will have been of a size sufficient to
receive any spoil or any prisoners worth the trouble of carrying back
again.
So far, however, as research has disclosed in those parts of the Near
East where civilisation was cradled, there is no indication that man
fought afloat—boat against boat, or fleet against fleet—until after a
comparatively high stage of civilisation had been attained and
shipbuilding had made enormous advances.
Evelyn remarks: “Concerning men of war, fleets, and armadas for
battel, that Minos was reported to be the author, which shows that
manner of desperate combat on the waters to be neer as antient as
men themselves, since the deluge.” Minos, he adds, disputed the
empire of the seas with Neptune, but “these particulars may be
uncertain.”[1]
Among the legendary expeditions, those of Ulysses and Jason are
the best known. Possibly they took place, but the adventurers never
did or saw half the wonders narrated of them. Herodotus, describing
the type of ship attributed to Ulysses by Homer, states that such
ships were made of acacia, of “planks about two cubits in length,”
joined together like bricks, and built in the following manner: “They
fasten the planks round stout and long ties: when they have thus
built the hulls they lay benches across them. They make no use of
ribs, but caulk the seams inside with byblus. They make only one
rudder, and that is driven through the keel. They use a mast of
acacia and sails of byblus. These vessels are unable to sail up the
stream, but are towed from the shore.”[2] Book II. of the Iliad
mentions, in the famous catalogue, hollow ships, well-benched
ships, swift ships, and dark ships, and that Ulysses had twelve red
ships, but Homer, being a poet and a landsman, did not describe
their differences.
Recent excavations and discoveries in Egypt have revealed the
existence of boats of considerable size, so remote in history that
their period is only guessed at, though they are estimated to date
from about 5000 to 6000 years b.c. If the interpretation of the
designs on the pottery recording these old ships be correct, they
were propelled by over a hundred oars or paddles, were steered by
three paddles at the stern, and had two cabins amidships. They
were, moreover, very high out of the water at the ends, having very
long, overhanging bows and counters, and were shallow and flat-
bottomed.
Even at this period the art of shipbuilding was in a comparatively
advanced stage; vessels such as those depicted would be quite as
capable of use in war for carrying warriors or stores, or both, as in
commerce for conveying merchandise. Egypt has many historical
secrets yet to reveal, and, judging by the constant reassignment of
dates in all matters connected with Ancient Egypt which exploration
has entailed, it is not too much to expect that the dates quoted,
assigned approximately by Egyptologists, may be revised and
events placed more remotely still.
Another hieroglyph, discovered in a tomb, ascribed to the year 4800
b.c., shows enormous progress in shipbuilding and also in the art of
representing a ship pictorially.
During the Sixth Dynasty, a certain Un’e, who was a person of note
under three kings, sent, while the second, Pepi I., was on the throne,
an expedition to the quarries of Syene or Assouan, to fetch stone for
his master’s pyramid.
Another expedition, on behalf of Pepi’s successor, merits attention,
as the fact is emphasised on the inscription on the tomb as most
remarkable, and as never having occurred before “under any king
whatever,” that Un’e had to employ twelve ships for freight and but
one warship.[3] The flotilla consisted of “six broad vessels, three tow
boats, three rafts, and one ship manned with warriors.”[4]
The Egyptians evidently had experience of some sort of fighting
afloat, for there has been discovered at Gebel Abu Faida a tomb with
a painting showing a boat with a triangular mast, and a stem
extending forward below the surface of the water and presumably
intended to be used to damage an enemy’s boat by ramming it.[5]
The first sea-fight of which a pictorial representation is known to exist
was fought off Migdol, at the mouth of the Nile, in the time of
Rameses III., first king of the Twentieth Dynasty, which began about
1180 b.c., and lasted to about 1050 b.c. Egypt was invaded from the
East by “warships and foot soldiers,” and the Egyptian monarch
mustered a fleet and attacked them.
ANCIENT EGYPTIAN WARSHIP.
“The ships on both sides,” says the historian[6]—“we can recognise

the Egyptian by the lion heads in the bows—have reefed their sails
in order not to interfere with the men who are fighting; the bracket at
the mast head has been removed to make room for the slinger. The
Egyptians understood how to pull round the ships of the enemy with
their grappling irons, so as to bring them to close quarters; in fighting
also they have the better of their opponents, for they all carry bows,
whilst the barbarians with their short swords can only fight in a hand-
to-hand medley. This battle is almost the only naval engagement in
Egyptian history, for though in the wars with the Hyksos we certainly
hear of fighting on the water, yet in the latter case the Nile was the
scene of action.... The ships had their individual names, such as
Battle Animal, or Glorious in Memphis. The Ship of Pharaoh was
also called Beloved of Amon.”
A remarkable difference between the ships of the Egyptians and
those of the Asiatics is that the latter had no rowers, if the bas-relief
is accurate. Possibly the Asiatics, Phoenicians probably, had
discovered how to manage the sails of their warships and dispense
with rowers.
The example set by the Asiatic fleet does not seem to have been
followed, for as the need of greater ships became manifest the
problem of their propulsion was met by placing one bank or tier of
oarsmen above another. Then, as now, the propelling power was
vital to the efficiency of the ship, and means had to be devised for
the preservation, or at least protection, of the oarsmen. The single-
banked ships had planks placed round the gunwales, forming a
parodus, or gangway, which served also to guard the rowers from
missiles. Later, the upper tier was in an open superstructure, and still
later, planks were carried which could be adjusted for the protection
of the oarsmen when necessary.
The ram, employed by the Egyptians—who seem to have retained
for their sea-going craft the long, overhanging stem and stern so
suitable to their river vessels—was a metal head, which added a
finishing touch to the projecting bows, and was high above the sea
level. At the time of the battle of Migdol, and possibly also of the sea-
fights in the reign of the preceding Rameses, who is known to have
conducted a naval war, though of this campaign no illustrations have
yet been discovered, the captain of the warship was placed in a sort
of crow’s nest on top of the double or ⋀-shaped mast.
Then comes a long gap in the history of Egyptian shipping. The
Phœnicians became the leading maritime people of the world, but
the little that is known of them is derived, not from discoveries in their
own cities of Tyre and Sidon, but from the records preserved at
Nineveh. Sennacherib’s conquest of Phœnicia was commemorated
by mural tablets, on which are the only known records of Phœnician
war galleys. The Phœnicians are stated to have invented biremes, or
vessels carrying two banks of oars on each side. Perhaps for
lightness, and in order to reduce the top weight as much as possible,
these galleys had the upper bank of oarsmen unprotected. The prow,
differing from that of the earlier Egyptian ships, curved forward at a
point slightly above the water line, and continued to do so under the
water, thus forming a formidable snout or ram which could inflict
considerable damage to the most vulnerable parts. The beaks were
generally carved to represent the head of some animal. The vessels
also had a parodus placed outside the vessel and extending the
whole length of the sides above the oars. The contrivance was
probably copied from the Egyptians, who introduced it to enable the
warriors to fight at close quarters when drawing alongside an enemy,
or to run to either end of the ship as occasion might require without
impeding, or being impeded by, the rowers.
GREEK BIREME.
From a Vase in the British Museum, found at Vulci.
GREEK WAR GALLEYS.
From a Vase in the British Museum found at Vulci.
Cancelli, or shields of basket work, were placed along the sides of
the ships at such a height that the heads of those on board are just
visible. The cancelli bore a striking resemblance to the circular
basket-work boats still to be found on the upper Euphrates; this
supports the supposition that the cancelli may have been used for
other purposes, particularly if they were made comparatively
watertight, as the function of a shield was not only to protect a
warrior in battle, but to help to keep him dry when on shipboard by
being disposed along the sides to prevent the spray from entering
the ships. A forecastle was constructed upon these ships, and upon
each forecastle a look-out man was stationed; and when these
structures came to be built of larger dimensions they served to
accommodate a number of fighting men who, from their superior
position, could throw their missiles with greater effect. The forecastle
had the further advantage of serving as a stronghold in the event of
an attempt being made to capture the ship by boarding it.
Following the Phœnicians, the Greeks are thought to have begun to
build their own warships about 700 b.c., perhaps earlier, but it was
about that time that the first three-banked warship was launched at
Corinth. The three-banked ships were for many years the largest in
existence. During the fourth century b.c. shipbuilding was practised
extensively, four-banked ships being built at Chalcedon, five-banked
at Salamis, and six-banked ships at Syracuse. Ships of ten banks,
according to Pliny, were ordered by Alexander the Great, and about
300 b.c. ships having twelve banks are said to have been built for
Ptolemy, and fifteen-banked ships for Demetrios, for a battle near
Cyprus.
Ptolemy Philopater, who ruled in Egypt from 222 to 204 b.c., is
alleged to have had a forty-banked ship of a length of 280 cubits or,
reckoning the cubit at 18 inches, of 420 feet, and a beam of 57 feet.
While increasing the size and number of oars, it would, nevertheless,
be impossible to augment to any appreciable extent the speed at
which these ships could be rowed, and the more unwieldy would
they become, and the more difficult would it be to keep steering way
upon them. Again, the assertions of the historians are so
contradictory that it is a thankless task to attempt to reconcile all their
stories, especially as they depended much upon hearsay for their
information. For that reason, therefore, a great deal that has been
recorded as to the early ships and their numerous banks of oars is
not to be accepted without careful inquiry and verification.
It has never been established beyond question what is meant by
banks of oars, or whether the Greek text has been interpreted
correctly when it is taken to express forty superimposed banks of
oars. From constructional reasons it may be assumed that a ship
having forty superimposed banks of oars never existed, and it is very
doubtful whether ships having more than a fourth of that number of
banks passed beyond the imaginations of their inventors. In any
case they were soon dispensed with, and in course of time it was
found that the best results were obtained with galleys having two or
three banks of oars.
It is not definitely known how the rowers were disposed in the ships
of anything over seven or eight banks. If any vessels had forty banks
of oars, the upper rows must have been of an absolutely unwieldy
length. Assuming the oars to have been weighted with lead so that
the inborne and outborne portions were equally balanced, they must
nevertheless have been exceedingly difficult to row even by a
number of men, and it was impossible for any rowers to have moved
these great oars at the same speed as the men at the lower banks
moved their lighter and shorter ones. That some such difficulty was
experienced, even in biremes and triremes, is shown by the
arrangement of the oars, whereby all in a bank were not of equal
length, but were graded so that those nearer the ends of the banks
were longer in order that all the blades might enter the water in a
straight line. Each row above must have had its own line in the water
a little farther away from the side of the ship than the row beneath it,
or the blades would have interfered with each other and the rowers
thrown into hopeless confusion. The tremendous amount of lead that
would have to be carried to counterbalance the outborne portions of
several hundred oars would add materially to the dead weight to be
propelled, and, much of it being placed high above the water, the
stability of the vessel would be lessened.
The Athenians used leather or skin aprons or covers over the oar
holes to prevent the water entering, the oar passing through a hole in
the leather, and the apron was bound to the oar in such a way as to
be watertight. This contrivance was widely adopted later. The oar
ports were constructed between the ribs, but the oars instead of
being rowed against the ribs were pulled against thongs fastened to
the next rib, thus minimising the strain upon the ship’s structure and
preventing the oars being lost overboard. One man one oar was
apparently the general rule at that time.
In his most painstaking study of “Ancient Ships” Mr. Cecil Torr has
gone very closely into the subject of the oar equipment of the
galleys. An Athenian three-banked ship would carry two hundred
oars, of which thirty were worked from the upper decking, sixty-two
on the upper bank, and fifty-four to each of the lower. The earliest
two-banked ships had eighteen rowers. An Athenian four-banked
ship might carry two hundred and sixty-six oars. The Roman and
Carthaginian five-banked ships in use about 256 b.c. had three
hundred rowers besides the combatants. The statement is made by
an early historian that in 280 b.c. the Heraclean fleet on the Black
Sea included an eight-banked ship with a hundred rowers on each
file, or one thousand six hundred rowers in all. As usually the fighting
men carried exceeded the rowers in number, the ship must have had
close upon three thousand five hundred men aboard.
Warships of all the early Eastern nations were strengthened by
cables passed longitudinally round them in order to keep the timbers
in place and prevent them from being started under the strain
occasioned by the shock of ramming. Egyptian ships of about 1200
b.c. had cables stretched from stem to stern and passing over the
top of the mast and other posts, but this contrivance was to prevent
the vessel from drooping at the ends, a weakness known as
“hogging.” The shock to the ramming vessel was scarcely less
severe than that to the vessel receiving the blow. To take up the
strain and add to the power of the blow the bows were strengthened
by means of waling pieces which supported the ram proper. The
Greek ships were built with the keel, the stempost, and the lower pair
of waling pieces converging to hold the ram, while higher up the
stem was a smaller ram which in its turn was buttressed by another
pair of waling planks. The catheads, or beams projecting from the
bows on either side by which the anchors were raised, were so
placed on a level with the gangway and gunwale that they would
sweep the upper works of an enemy’s ship and smash its gangway
and hurl into the sea or the hold all the fighting men upon it. Ships of
more than three banks are believed to have carried another ram
level with the catheads, and to have had a ram for every pair of
additional waling beams. The ram heads were generally of bronze
and weighed 170 lb. or more.
“AN ANCIENT BIREME, FROM BASIUS, HAVING ONE TIER OF OARS ONLY.”
“ONE OF THE ANCIENT LIBURNI, OR GALLEYS, HAVING A SINGLE TIER

OF OARS, ACCORDING TO BASIUS.”
AN ANCIENT TRIREME, ACCORDING TO BASIUS.
From Charnock’s “History of the Marine Architecture of all Nations.”
The later rams varied considerably in shape. The triple ram was
sometimes made with the teeth pointing slightly downward, while
others had an upward tilt. The lowest ram often extended farther
forward than those above, the idea being that it would inflict severe
injury about or below the water line, and that the upper rams,
besides causing damage, would push the stricken vessel off the
lower ram and let her sink without the assailant being dragged down
by the head with her.
The build of the ships rendered it necessary that an engagement
should be fought on a calm sea, and daylight was preferred in order
that the combatants could see what they were doing. As the fleets
approached one another the commanders of the different vessels
decided upon their individual opponents. Much skilful manœuvring
ensued to ram the enemy or avoid a blow. The slaves strained at the
oars while their taskmasters ran between the files of rowers and,
with unmerciful blows from heavy sticks and whips, stimulated them
to still greater exertions if possible.
Poor slaves, mostly prisoners of war, their prospects were gloomy in
the extreme! If their ship were rammed some of them were sure to
be injured, and if she sank they went down with her, fastened to their
places and having no chance of escape. If the oars were disabled in
the collision between the ships the rowers were bound to receive
violent blows from the inboard end of the oars, or to be cruelly
pierced by splinters of wreckage. Showers of missiles from the
opposing ship fell upon the helpless wretches. In later years, when
the terrible Greek fire was added to the means of attack and
defence, it contributed the prospect of being burnt alive to the other
horrors of their situation. Victory meant no rejoicings for them. The
wounded were of little account and could be dispensed with when
slaves were to be had for the capturing, and it was easy to put them
overboard to die the more quickly. Those who survived the battle
unhurt or not too severely injured to recover rapidly, were retained. If
their ship were vanquished they might look forward to greater
cruelties as a punishment for their share of the defeat. If they
belonged to the victors, they had only more battles, the torturing
whips of their drivers, and insufficient food as their portion in life.
Death came as a welcome relief to the slaves of victor and
vanquished; in it lay their only hope of peace.
When the Roman navy was at its best the ships were painted a
colour which matched the waves, and the hulls were made as
watertight as possible with tar. Occasionally in the later Roman ships
layers of tarred cloth were placed outside the outer planking, and the
hull was then lead-sheathed. Bronze nails and wooden pegs were
used in fastening the timbers together, and some ships were so built
that they could be taken to pieces and transported overland if
necessary. Ships of three, four and five banks were even conveyed
from the Mediterranean to the Euphrates.
The facility with which the Liburnians handled the two-banked ships
in their Adriatic campaign induced the Romans to adopt these
vessels as models for their own two-banked ships, and in course of
time they adopted the name of liburna for all war-vessels of from one
to five banks.
If some of the historians may be believed, anything that could be
piled upon the ancient ship and did not capsize it was permissible.
One is said to have had a tower at the stern and another at the prow.
Another bore “a large tower of masonry with a great gate. Here
appear some vases, probably filled with combustibles.” Another
libernus has a mast or yard, suspended perpendicularly by the side
of the forward tower, and having at each end a crossbeam. Yet
another libernus, besides carrying a protector for the helm at the
stern, is said to have had six round towers; the largest, of embattled
masonry, was at the prow, two others, also of masonry, surmounted
by domes, and connected by a bridge, were near the stern, and the
other three were nearer the fore part of the ship, were roofed, and
two of them had windows.
Shipping in the Mediterranean extended with extraordinary rapidity in
the recovery after the stagnation caused by the fall of the Roman
Empire and the relapse into semi-barbarism which followed the
successful invasion of Italy by the wild tribesmen of the North. The
advent and rise of the Moslem power caused a series of struggles in
which every state was in a more or less constant condition of warfare
against its neighbour, and the Crusades served but to add fuel to the
fire of internecine and religious conflict. Some immense ships are
stated to have been employed up to and at the fall of Constantinople.
The early centuries of the Christian era saw the evolution of a flat,
shallow vessel, fitted with one or two masts carrying sails, from
which the lateen rig developed, equipped with a long ram above the
water line, with two or at most three banks of oars. It appears from
illustrations that some of these boats carried a superstructure
extending beyond the beam on either side. War vessels of this type
became common throughout the length and breadth of the
Mediterranean, and remained in use long after the introduction of
firearms.
Before the discovery of Greek fire, flaring missiles of some kind had
been devised. Frontinus mentions fire-ships, or hulls carrying
combustibles and allowed to drift with wind and tide upon the
enemy’s ships: stinkpots, to nauseate the enemy, though how the
others escaped the smells except by keeping to windward does not
appear; and Evelyn adds, “Nay, snake pots, and false colours.” The
Greek fire, however, was the most terrible of the weapons employed
at that time. By some means by which a fair amount of power was
exerted, the liquid was squirted—or vomited, to use one historian’s
phrase—through copper pipes upon an enemy’s ship, and as the
liquid had the peculiar property of igniting upon exposure to the air
and was inextinguishable by water, it was a most formidable engine
of destruction. Small vases filled with the liquid and sealed airtight
were used as hand grenades and flung at opposing ships and,
breaking, set them on fire. Heavy arrows carrying balls of flax
soaked in the liquid were used both in land and sea warfare, as also
were hand-flung javelins similarly equipped, and the flights of these
masses of inextinguishable flames must have been equally
demoralising to the combatants against whom they were directed
and destructive to the ships and inflammable buildings upon which
they fell. This composition is thought to have been invented in the
seventh century; the first occasion on which it was employed on an
extensive scale was in the great battle between the fleets of
Constantine and the Saracens, when the latter, through its agency,
lost practically their whole fleet and thirty thousand men killed. After
that both sides used Greek fire whenever possible.
Up to the introduction of gunpowder and artillery the methods of
fighting varied but little. The sea-fights of the Crusades were
conducted on the lines which had been recognised as the best for a
couple of thousand years or more, viz., ram the enemy and board
him. Greek fire added this rule: Burn him also if you can.
The countries along the northern and southern shores of the
Mediterranean had attained a high degree of civilisation when the
inhabitants of Western Europe and the British Islands were still more
or less savage. What may be regarded as circumstantial evidence in
support of the contention that the Phœnicians voyaged to Cornwall
and Ireland is the similarity which exists in shape between the wicker
shields, such as the Phœnicians are known to have used, and the
wicker coracles which the Britons employed at the time of the
invasion by Julius Cæsar. There must have been considerable
intercourse between the Phœnicians and the dwellers in the valley of
the Euphrates before the latter conquered the former; but whether
the dwellers in Nineveh, or those by the sea, invented wicker boats,
or whether both derived their knowledge of wicker boats from other
sources, are points of no immediate importance. But what is of
interest is that the British wicker coracles were covered with hides to
make them watertight, that they had keels and gunwales, and that
they were small enough to be used as shields if necessary, their
dimensions being rather over 4 feet in length, with a breadth of about
3 feet, and a depth of a trifle over 12 inches. They were big enough
to carry one man of average size. There are on the Euphrates to this
day boats or rafts of proportionate dimensions, up to a maximum
length of 40 or 50 feet over all, which are constructed with a light
framework of wicker and timber, over which skins are stretched to
keep them watertight. These boats, when laden, drift down the river
with the current, and, on reaching their destination, their cargo and
skins are sold and the framework is made up into a package and
returned upon the back of an ass to the port of departure. These
cargo boats have been humorously referred to at a meeting of the
Institute of Marine Engineers as of “one ass-power.”
So far as Britain is concerned, the shipping of each coast seems to
have developed under the influence of the foreign shipping with
which it mostly came in contact. The east coast was largely
concerned with the Danes, and the south coast with its neighbours
across the Channel. The Danes and Vikings developed a type of
vessel peculiarly their own. The best specimen yet brought to light is
that known as the Gokstad ship.

Inverse Synthetic Aperture Radar Imaging With Matlab Algorithms 2Nd Edition Caner Ozdemir Full Chapter PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Inverse Synthetic Aperture Radar Imaging With Matlab Algorithms 2Nd Edition Caner Ozdemir Full Chapter PDF

Uploaded by

Copyright:

Available Formats

Inverse Synthetic Aperture Radar

Imaging With MATLAB Algorithms 2nd

Advanced Algorithms for Mineral and Hydrocarbon

Nonlinear Ocean Dynamics: Synthetic Aperture Radar 1st

Introduction to Synthetic Aperture Radar: Concepts and

Modern Information Optics with MATLAB Yaping Zhang

Fundamental Chemistry with MATLAB 1st Edition Daniele

Signal processing for multistatic radar systems:

Data Structures and Algorithms in C++ 2nd Edition

Cryptography Algorithms - 2nd Edition (Early Release)

With Advanced SAR/ISAR Imaging Concepts,

Caner Özdemir, Phd

Preface to the Second Edition xvi

1 Basics of Fourier Analysis 1

1.4 Convolution and Multiplication Using FT 11

3 Synthetic Aperture Radar 85

3.11 Advanced Topics in SAR 140

5 Imaging Issues in Inverse Synthetic Aperture Radar 246

5.4 Zero Padding 260

6 Range-Doppler Inverse Synthetic Aperture Radar Processing 306

6.8.5 Pulse Repetition Frequency 329

7 Scattering Center Representation of Inverse Synthetic Aperture

8 Motion Compensation for Inverse Synthetic Aperture Radar 385

8.3.4 Algorithm for JTF-Based Translational and Rotational MOCOMP 408

9 Bistatic ISAR Imaging 440

10 Polarimetric ISAR Imaging 484

10.4 ISAR Imaging with Full Polarization 497

11 Near-Field ISAR Imaging 533

11.5.3 Near-Field ISAR Imaging of a Small Object 555

12 Some Imaging Applications Based on SAR/ISAR 571

Preface to the Second Edition

other SAR focusing/processing algorithms including chirp scaling algorithm

• In Chapters 4, 5, 7, and 12 where we handle various aspects of ISAR imaging

• In Chapter 12 in which some examples based on SAR/ISAR imaging technol-

Mersin, October 2020 Caner Özdemir

PRI Pulse repetition interval

Basics of Fourier Analysis

1.1 Forward and Inverse Fourier Transform

1.1.1 Brief History of FT

of the computational time of the transformation operation (Runge 1903). In 1942,

1.1.2 Forward FT Operation

1.2 FT Rules and Pairs

1.2.2 Time Shifting

1.2.3 Frequency Shifting

1.2.6 Time Reversal

1.2.11 Derivation and Integration

1.2.12 Parseval’s Relationship

1.3 Time-Frequency Representation of a Signal

1.3.1 Signal in the Time Domain

1.3.2 Signal in the Frequency Domain

Time domain signal

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

Figure 1.1 The time-domain signal of “prince” spoken by a lady.

Frequency domain signal

Figure 1.2 The frequency-domain signal (or the spectrum) of “prince.”

advantageous than the time-domain representation. The frequency-domain signal

1.3.3 Signal in the Joint Time-Frequency (JTF) Plane

Signal in time-frequency plane

Figure 1.3 The time-frequency representation of the word “prince.”

1.4 Convolution and Multiplication Using FT

In a dual manner, convolution between the frequency-domain signals can be

Figure 1.6 An ideal and real LP filter characteristics.

achieve an ideal band-limited characteristic as in Figure 1.6, then the correspond-