Bellingham, Washington USA
Library of Congress CataloginginPublication Data
Advances in information optics and photonics / Ari T. Friberg and René Dändliker. p. cm.  (Press monograph ; PM183) Includes bibliographical references and index. ISBN 9780819472342 1. Optical communications. 2. Photonics. I. Friberg, Ari T., 1951 II. Dändliker, René. TK5103.59.A35 2008
621.382'7dc22
Published by
2008022953
SPIE P.O. Box 10 Bellingham, Washington 982270010 USA Phone: +1 360 676 3290 Fax: +1 360 647 1445 Email: spie@spie.org Web: http://spie.org
Copyright © 2008 Society of PhotoOptical Instrumentation Engineers
All rights reserved. No part of this publication may be reproduced or distributed in any form or by any means without written permission of the publisher.
The content of this book reflects the work and thought of the author(s). Every effort has been made to publish reliable and accurate information herein, but the publisher is not responsible for the validity of the information or for any outcomes resulting from reliance thereon.
Printed in the United States of America.
Contents
List of Contributors 
ix 
Preface 
xv 
ICO International Trends in Optics Series History 
xix 
I. Beam Optics 

1. FirstOrder Optical Systems for Information Processing Tatiana Alieva 
1 
2. Applications of the Wigner Distribution to Partially Coherent Light Beams Martin J. Bastiaans 
27 
3. Characterization of Elliptic Dark Hollow Beams Julio C. GutiérrezVega 
57 
4. Transfer of Information Using Helically Phased Modes Miles Padgett, Graham Gibson, and Johannes Courtial 
77 
II. Laser Photonics and Components 

5. Microoptical Components for Information Optics and Photonics Christof Debaes, Heidi Ottevaere, and Hugo Thienpont 
89 
6. Intracavity Coherent Addition of Lasers Vardit Eckhouse, Amiel A. Ishaaya, Liran Shimshi, Nir Davidson, and Asher A. Friesem 
117 
7. Light Confinement in Photonic Crystal Microcavities Philippe Lalanne and Christophe Sauvan 
137 
v
vi
Contents
8. 
Limits to Optical Components David A. B. Miller 
153 
III. Electromagnetic Coherence 

9. An Overview of Coherence and Polarization Properties for Multicomponent Electromagnetic Waves Alfredo Luis 
171 

10. Intrinsic Degrees of Coherence for Electromagnetic Fields Philippe Réfrégier and Antoine Roueff 
189 

IV. Imaging, Microscopy, Holography, and Materials 

11. Digital Computational Imaging Leonid Yaroslavsky 
209 

12. Superresolution Processing of the Response in Scanning Differential Heterodyne Microscopy Dmitry V. Baranov and Evgeny M. Zolotov 
229 

13. Fourier Holography Techniques for Artificial Intelligence Alexander V. Pavlov 
251 

14. Division of Recording Plane for Multiple Recording and Its Digital Reconstruction Based on Fourier Optics Guoguang Mu and Hongchen Zhai 
271 

15. Fundamentals and Advances in Holographic Materials for Optical Data Storage Maria L. Calvo and Pavel Cheben 
285 

16. Holographic Data Storage in LowShrinkage Doped Photopolymer 
317 
Shiuan Huei Lin, Matthias Gruber, YiNan Hsiao, and Ken Y. Hsu
Contents
vii
V. Photonic Processing
17. Temporal Optical Processing Based on Talbot’s Effects Jürgen Jahns, Adolf W. Lohmann, and Hans Knuppertz 
343 
18. Spectral LinebyLine Shaping Andrew M. Weiner, ChenBin Huang, Zhi Jiang, Daniel E. Leaird, and Jose Caraquitena 
359 
19. Optical Processing with Longitudinally Decomposed Ultrashort Optical Pulses Robert Saperstein and Yeshaiahu Fainman 
381 
20. Ultrafast Information Transmission by QuasiDiscrete Spectral Supercontinuum Mikhail A. Bakhtin, Victor G. Bespalov, Vitali N. Krylov, Yuri A. Shpolyanskiy, and Sergei A. Kozlov 
405 
VI. Quantum Information and Matter 

21. Noise in Classical and Quantum PhotonCorrelation Imaging Bahaa E. A. Saleh and Malvin Carl Teich 
423 
22. Spectral and Correlation Properties of TwoPhoton Light Maria V. Chekhova 
437 
23. EntanglementBased Quantum Communication Alexios Beveratos and Sébastien Tanzilli 
457 
24. Exploiting Optomechanical Interactions in Quantum Information Claudiu Genes, David Vitali, and Paolo Tombesi 
489 
25. Optimal Approximation of NonPhysical Maps via Maximum Likelihood Estimation Vladimír Bužek, Mário Ziman, and Martin Plesch 
513 
26.
Quantum Processing Photonic States in Optical Lattices Christine A. Muschik, Inés de Vega, Diego Porras, and J. Ignacio Cirac
533
viii
Contents
27. 
Strongly Correlated Quantum Phases of Ultracold Atoms in 

Optical Lattices Immanuel Bloch 
555 

VII. Communications and Networks 

28. The Intimate Integration of Photonics and Electronics Ashok V. Krishnamoorthy 
581 

29. Echelle and Arrayed Waveguide Gratings for WDM and Spectral Analysis Pavel Cheben, André Delâge, Siegfried Janz, and DanXia Xu 
599 

30. Silicon Photonics—Recent Advances in Device Development Andrew P. Knights and J. K. Doylend 
633 

31. Toward Photonic Integrated Circuit AllOptical Signal Processing Base on Kerr Nonlinearities David J. Moss and Benjamin J. Eggleton 
657 

32. Ultrafast Photonic Processing Applied to Photonic Networks Hideyuki Sotobayashi 
687 

Index 
713 
List of Contributors
Tatiana Alieva Universidad Complutense de Madrid, Spain
Mikhail A. Bakhtin State University of Information Technologies, Russia
Dmitry V. Baranov General Physics Institute of Russian Academy of Sciences, Russia
Martin J. Bastiaans Technische Universiteit Eindhoven, The Netherlands
Victor G. Bespalov State University of Information Technologies, Russia
Alexios Beveratos Alcatel de Marcoussis, France
Immanuel Bloch Johannes GutenbergUniversität, Germany
Vladimír Bužek Research Center for Quantum Information and QUNIVERSE, Slovakia
María L. Calvo Universidad Complutense de Madrid, Spain
Jose Caraquitena Purdue University, USA
ix
x
List of Contributors
Pavel Cheben National Research Council Canada
Pavel Cheben National Research Council Canada
Maria V. Chekhova M.V. Lomonosov Moscow State University, Russia
J. Ignacio Cirac
MaxPlanckInstitut für Quantenoptik, Germany
Johannes Courtial University of Glasgow, Scotland
Nir Davidson Weizmann Institute of Science, Israel
Christof Debaes Vrije Universiteit Brussel, Belgium
André Delâge National Research Council Canada
J. K. Doylend
McMaster University, Canada
Vardit Eckhouse Weizmann Institute of Science, Israel
Benjamin J. Eggleton University of Sydney, Australia
Yeshaiahu Fainman University of California San Diego, USA
Asher A. Friesem Weizmann Institute of Science, Israel
Claudiu Genes Università di Camerino, Italy
List of Contributors
xi
Graham Gibson University of Glasgow, Scotland
Matthias Gruber Fern Universität Hagen, Germany
Julio C. GutiérrezVega Optics Center Tecnológico de Monterrey, México
YiNan Hsiao National Chiao Tung University, Taiwan
Ken Y. Hsu National Chiao Tung University, Taiwan
ChenBin Huang Purdue University, USA
Amiel A. Ishaaya Weizmann Institute of Science, Israel
Jürgen Jahns FernUniversität Hagen, Germany
Siegfried Janz National Research Council Canada
Zhi Jiang Purdue University, USA
Andrew P. Knights McMaster University, Canada
Hans Knuppertz FernUniversität Hagen, Germany
Sergei A. Kozlov State University of Information Technologies, Russia
Ashok V. Krishnamoorthy Sun Microsystems, USA
xii
List of Contributors
Vitali N. Krylov State University of Information Technologies, Russia
Philippe Lalanne Université ParisSud, France
Daniel E. Leaird Purdue University, USA
Shiuan Huei Lin National Chiao Tung University, Taiwan
Adolf W Lohmann University of Erlangen, Germany
Alfredo Luis Universidad Complutense, Spain
David A. B. Miller Stanford University, USA
David J. Moss University of Sydney, Australia
Guoguang Mu Nankai University, China
Christine A. Muschik MaxPlanckInstitut für Quantenoptik, Germany
Heidi Ottevaere Vrije Universiteit Brussel, Belgium
Miles Padgett University of Glasgow, Scotland
Alexander V. Pavlov St. Petersburg State University for Information Technologies, Russia
List of Contributors
xiii
Martin Plesch Research Center for Quantum Information and QUNIVERSE, Slovakia
Diego Porras MaxPlanckInstitut für Quantenoptik, Germany
Philippe Réfrégier Institut Fresnel, AixMarseille Université, France
Antoine Roueff Institut Fresnel, AixMarseille Université, France
Bahaa E. A. Saleh Boston University, USA
Robert Saperstein University of California San Diego, USA
Christophe Sauvan Université ParisSud, France
Liran Shimshi Weizmann Institute of Science, Israel
Yuri A. Shpolyanskiy State University of Information Technologies, Russia
Hideyuki Sotobayashi Aoyama Gakuin University and National Institute of Information and Communications Technology, Japan
Sébastien Tanzilli Université de Nice SophiaAntipolis, France
Malvin Carl Teich Boston University, USA
Hugo Thienpont Vrije Universiteit Brussel, Belgium
xiv
List of Contributors
Paolo Tombesi Università di Camerino, Italy
Inés de Vega MaxPlanckInstitut für Quantenoptik, Germany
David Vitali Università di Camerino, Italy
Andrew M. Weiner Purdue University, USA
DanXia Xu National Research Council Canada
Leonid Yaroslavsky Tel Aviv University, Israel
Hongchen Zhai Nankai University, China
Mário Ziman Research Center for Quantum Information and QUNIVERSE, Slovakia
Evgeny M. Zolotov General Physics Institute of Russian Academy of Sciences, Russia
Preface
This volume is the sixth in a series of books that the International Commission for Optics (ICO) edits for publication at the time of its triennial congresses. The earlier volumes have covered a broad scope of interests in optics at the time and have dealt with fundamental subjects, while the later editions have increasingly addressed advances in applied optics and photonics. The books previously published in the series are
• International Trends in Optics, ed. J. W. Goodman, USA (Academic Press, 1991)
• Current Trends in Optics, ed. J. C. Dainty, UK (Academic Press,
1994)
• Trends in Optics – Research, Developments and Applications, ed. A. Consortini, Italy (Academic Press, 1996)
• International Trends in Optics and Photonics, ed. T. Asakura, Japan (Springer, 1999)
• International Trends in Applied Optics, ed. A. H. Guenther, USA (SPIE Press, 2002)
The complete history of the ICO Book series, including the Tables of Contents of the previous volumes, can be found on p. xix of this book. Besides highlighting the main developments of international optics and photonics, the aim of this book series is to promote the general awareness of the ICO and raise funds for its global activities, in particular the travelling lecturer program, which is aimed at enhancing optics in developing nations. Therefore all royalties will go to the ICO for that purpose. In today’s ‘age of light,’ optical information science and technology play a central role. The ICO has a long tradition in the subjects of information optics, dating back to the ICO topical meetings in Kyoto, Japan 1994 (Frontiers in Information Optics) and Tianjin, China 1998 (Optics for Information Infrastructure). The ICO has also been a permanent sponsor of the Optical Computing/Optics in Computing conferences, a series of meetings spanning well over a decade. In 2006, the ICO organized two key events on information optics: the ICO topical
xv
xvi
Preface
meeting on Optoinformatics / Information Photonics in St. Petersburg, Russia (Chairs A. V. Pavlov, M. L. Calvo, and J. Jahns) and the ICO/ICTP Winter College on Optics in Trieste, Italy, with title “Quantum and Classical Aspects of Information Optics” (Directors P. Tombesi, M. L. Calvo, and P. Knight). Additionally, the recent ICO Prizes – most notably those in 2003 (B. J. Eggleton), 2004 (A. V. Krishnamoorthy), 2005 (I. Bloch), and 2006 (H. Sotobayashi) – have dealt with various basic and applied aspects of optical information. Hence it was quite natural to take
advantage of these developments and focus the current volume of the ICO Book series on Advances in Information Optics and Photonics. The present volume VI differs from the previous ones in at least three respects: it concentrates on a specific, though extremely important, topic within the broad field of optics and photonics, it does not contain the words ‘International Trends’ explicitly in the title, and it is published as a paperback. We hope that with these changes the book will find its way as
a standard reading and reference material on the topic. The volume
consists of 32 invited contributions from scientists or research groups working throughout the world on optical information science, technology, and applications. Many of the authors have actively participated in the ICO conferences and other activities and all of them are internationally recognized leaders in their respective subjects.
Many new concepts in classical and quantumentangled light, coherent interaction with matter, novel materials and processes have led to remarkable breakthroughs in information science and technology. While it
is difficult, and sometimes even dangerous, to group the contributions
under separate headings, we have divided the chapters of this book into 7 sections:
1. Beam Optics
2. Laser Photonics and Components
3. Electromagnetic Coherence
4. Imaging, Microscopy, Holography, and Materials
5. Photonic Processing
6. Quantum Information and Matter
7. Communications and Networks
The sections contain chapters that address optical information sciences
broadly in the linear, nonlinear, classical, and quantum regimes and describe the foundations, stateoftheart devices and technologies, as well
as the diverse applications of information optics and photonics. It is hoped
that the reader will find chapters that are directly relevant to his/her own
Preface
xvii
work or otherwise will create interest in this fascinating, rapidly advancing, and highly potential subject. We would like to express our sincere appreciation to all of the authors who have devoted their time, effort, and expertise to write the superb and timely contributions for this volume. We would also like to thank the staff of SPIE Press, and especially Merry Schnell, Gwen Weerts, and Eric Pepper, for their professional work to produce this highquality publication for the benefit of the global optics and photonics community.
Ari T. Friberg President, International Commission for Optics Royal Institute of Technology (KTH), Stockholm, Sweden Helsinki University of Technology (TKK), Espoo, Finland University of Joensuu, Finland
René Dändliker Past President, International Commission for Optics President of the Swiss Academy of Engineering Sciences University of Neuchâtel, Switzerland
ICO International Trends in Optics Series History
The first book in the series appeared in 1991 under the title “International Trends in Optics” in the Lasers and Optical Engineering series of Academic Press. The Editor was 1987–1990 ICO President, Prof. J.W. Goodman of Stanford University, USA. It includes the following chapters:
• Integrated Optics, OEICs, or PICs? H. Kogelnik
• Quantum Optoelectronics for Optical Processing, D.A.B. Miller
Optics • in 
Telecommunications: 
Beyond 
Transmission, 
P.W.E. 
Smith 
• Microoptics, Kenichi Iga
• Holographic Optical Elements for Use with Semiconductor Lasers, H.P. Herzig and R. Dändliker
• FibreOptic Signal Processing, B. Culshaw, I. Andonovic
• Optical Memories, Yoshito Tsunoda
• How Can Photorefractives Be Used? H. Rajbenbach and J.P. Huignard
• Adaptice Interferometry: A New Area of Applications of Photorefractive Crystals, S.I. Stepanov
• Water Wave Optics, J. J. Stamnes
• About the Philosophies of Diffraction, A.W. Lohmann
• The Essential Journals of Optics, J.N. Howard
• Optics in China: Ancient and Modern Accomplishments, Z.M. Zhang
• Unusual Optics: Optical Interconnects as Learned from the Eyes of Nocturnal Insects, Crayfish, Shellfish, and Similar Creatures, P. Greguss
• The Opposition Effect in Volume and Surface Scattering, J.C. Dainty
xix
ICO International Trends in Optics Series History
xx
• Influence of Source Correlations on Spectra of Radiated Fields, E. Wolf
• Quantum Statistics and Coherence of Nonlinear Optical Processes,
J. Peřina
• OnePhoton Light Pulses versus Attenuated Classical Light Pulses,
A. Aspect and P. Grangier
• Optical Propagation through the Atmosphere, A. Consortini
• Are the Fundamental Principles of Holography Sufficient for the Creation of New Types of 3D Cinematography and Artificial Intelligence? Y. Denisyuk
• Medical Applications of Holographic 3D Display, J. Tsujiuchi
• Moiré Fringes and Their Applications, O. Bryngdahl
• Breaking the Boundaries of Optical System Design and Construction, C.H.F. Velzel
• Interferometry: What’s New Since Michelson? P. Hariharan
• Current Trends in Optical Testing, D. Malacara
• Adaptive Optics, F. Merkle
• Triple Correlations and Bispectra in HighResolution Astronomical Imaging, G. Weigelt
• PhaseRetrieval Imaging Problems, J.R. Fienup
• Blind Deconvolution—Recovering the Seemingly Irrecoverable! R.H.T Bates and H. Jiang
• Pattern Recognition, Similarity, Neural Nets, and Optics, H.H. Arsenault, Y. Sheng
• Towards
Nonlinear
Optical
Processing,
T.
Szoplik
and
K.
ChalasinskaMacukow
• New Aspects of Optics for Optical Computing, V. Morozov
• Digital Optical Computing, S.D. Smith and E.W. Martin
• Computing: A Joint Venture for Light and Electricity? P. Chavel
The second book in the series appeared under the title “Current Trends in Optics” in the Lasers and Optical Engineering series of Academic Press Limited, London, 1994 (ISBN 012207204). The Editor is ICO Past President, Prof. J.C. Dainty of Imperial College, London. It includes the following chapters:
• Atomic Optics, S.M. Tan and D.F. Walls
• Single Atoms in Cavities and Traps, H. Walther
xxi
ICO International Trends in Optics Series History
• Meet a Squeezed State and Interfere in Phase Space, D. Kr hmer,
E. Mayr, K. Vogel and W.P. Schleich
• Can Light Be Localized? A. Lagendijk
• Timeresolved Laserinduced Breakdown Spectrometry, G. Lupkovics, B. Nemet and L. Kozma
• Fractal Optics, J. Uozumi and T. Asakura
• On the Spatial Parametric Characterization of General Light Beams, R. MartinezHerrero and P.M. Mejias
• To See the Unseen: Vision in Scattering Media, E.P. Zege and I.L. Katsev
• Backscattering
Through
Turbulence,
A.S.
Gurvich
and
A.N.
Bogaturov
• Why is the Fresnel Transform So Little Known? F. Gori
• Fourier Curios, A.W. Lohmann
• The Future of Optical Correlators, D. Casasent
• Spectral Hole Burning and Optical Information Processing, K.K. Rebane
• Holographic Storage Revisited, G.T. Sincerbox
• Colour Information in Optical Pattern Recognition, M.J. Yzuel and
J. Campos
• The Optics of Confocal Microscopy, C.J.R. Sheppard
• Diffraction Unlimited Optics, A. Lewis
• Superresolution in Microscopy, V.P. Tychinsky and C.H.F. Velzel
• Fringe Analysis: Anything New? M. Kujawinska
• Diagnosing the Aberrations of the Hubble Space Telescope, J.R. Fienup
• Laser Beacon Adaptive Optics:Boom or Bust? R.Q. Fugate
The third book in the series appeared in August 1996 under the title Trends in Optics—Research, Developments and Applications, ISBN 012 1860302. Like its two predecessors, it was published by Academic Press. The Editor is ICO Past President, Prof. Anna Consortini of Universita degli Studi di Firenze, Italy. The Museo ed Istituto della Scienza in Florence deserves thanks for its permission to use the photography of one of its Galileo Galilei lenses as cover illustration. The book includes the following chapters:
• A Short History of the Optics Group of the Willow Run Laboratories, E.N. Leith
• Biospeckles, Y. Aizu and T. Asakura
ICO International Trends in Optics Series History
xxii
• Photon Migration and Imaging of Biological Tissues, G. Zaccanti and D. Contini
• Direct Image Processing Using Artificial Retina Chips, E. Lange,
Y. Nitta and K. Kyuma
• Principles and Development of Diffraction Tomography, E. Wolf
• Diffractive Optics: From Promise to Fruition, J. Turunen and F. Wyrowski
• Planar Diffractive Elements for Compact Optics, A.A. Friesem and
Y. Amitai
• Resonant Light Scattering from Weakly Rough Metal Surfaces, K.A. O'Donnell
• Femtosecond TimeandSpaceDomain Holography, A. Rebane
• Holographic 3D disks Using Shift Multiplexing, D. Psaltis, G. Barbastathis and M. Levene
• Dense Optical Interconnections for Silicon Electronics, D.A.B. Miller
• Fanin Loss for Electrical and Optical Interconnections, J.W. Goodman and J.C. Lain
• Signal Processing and Storage Using Hybrid ElectroOptical Procedures, J. Shamir
• Young’s Experiment in Signal Synthesis, J. OjedaCastaeda and A.W. Lohmann
• Resolution Enhancement by Data Inversion Techniques, C. de Mol
• Electronic Speckle Pattern Interferometry: An Aid in Cultural Heritage Protection, G. Schirripa Spagnolo
• Numerical Simulation of Irradiance Fluctuations for Optical Waves Through Atmospheric Turbulence, S.M. Flatte
• Optical Scintillation Methods of Measuring Atmospheric Surface Fluxes of Heat and Momentum, R.J. Hill
• Coherent Doppler Lidar Measurements of Winds, R. Frehlich
• Doing Coherent Optics with Soft XRay Sources, D. Joyeux, P. Jaegle and A. l'Huillier
• Axially Symmetric Multiple Mirror Optics for Soft XRay Projection Microlithography, S.S. Lee, C.S. Rim, Y.M. Cho, D.E. Kim and C.H. Nam
• Olmec Mirrors: An Example of Archaeological American Mirrors, J.J. Lunazzi
• Galileo Galilei: Research and Development of the Telescope, G. Molesini and V. Greco
xxiii
ICO International Trends in Optics Series History
• GRIN Optics: Practical Elements, C. GomezReino and J. Linares Beiras
• Photorefractive Fibers: Fabrication and Hologram Construction, F.T.S. Yu and S. Yin
• Optical Morphogenesis: Dynamics of Patterns in Passive Optical Systems, F.T. Arecchi, S. Boccaletti, E. Pampaloni, P.L. Ramazza and S. Residori
• High Sensitivity Molecular Spectroscopy with Diode Lasers, K. Ernst
• Submicrometre Optical Metrology Using Laser Diodes and Polychromatic Light Sources, C. Gorecki and P. Sandoz
• A Physical Method for Colour Photography, G.G. Mu, Z.L. Fang, F.L. Liu and H.C. Zhai
• Multiwavelength Vertical Cavity Laser Arrays by Molecular Beam Epitaxy, C.J. ChangHasnain, W. Yeun, G.S. Li and L.E. Eng
• Compact BlueGreen Laser Sources, W.J. Kozlovsy
The fourth book in the series appeared in August 1999 under the title International Trends in Optics and Photonics. It was editored by Prof. T. Asakura and published by SpringerVerlag as Volume 74 of the Springer Series in Optical Sciences. The book includes the following chapters:
• Optical Twist, A.T. Friberg
• Principles and Fundamentals of Near Field Optics, M. Nieto Vesperinas
• SpinOrbit Interaction of a Photon: Theory and Experiment on the Mutual Influence of Polarization and Propagation, N.D. Kundikova and B.Ya. Zel'dovich
• Atoms and Cavities: the Birth of a Schroedinger Cat of the Radiation Field, J.M. Raimond and S. Haroche
• Quantum Tomography of Wigner Functions from Incomplete Data, V. Buzek, JG. Drobny and H. Wiedemann
• Some New Aspects on the Resolution in Gaussian Pupil Optics, S.S. Lee, M.H. Lee and Y.R. Song
• Multichannel Photography with Digital Fourier Optics, G.G. Mu, L. Lin and Z.Q. Wang
• Holographic Optics for Beamsplitting and Image Multiplication, A.L. Mikaelian, A.N. Palagushkin and S.A. Prokopenko
• Image Restoration, Enhancement and Target Location with Local Adaptive Linear Filters, L. Yaroslavsky
ICO International Trends in Optics Series History
xxiv
• Fuzzy Problem for Correlation Recognition in Optical Digital Image Processing, G. Cheng, G. Jin, M. Wu and Y. Yan
• AllOptical Regeneration for GlobalDistance FiberOptic Communications, E. Desurvire and O. Leclerc
• Non Quantum Cryptography for Secure Optical Communications, J.P. Goedgebuer
• Pulsed Laser Deposition: An Overview, I.N. Mihailescu and E. Gyorgy
• Absolute Scale of Quadratic NonlinearOptical Susceptibilities, I. Shoji, T. Kondo and R. Ito
• Femtosecond Fourier Optics: Shaping and Processing of Ultrashort, Optical Pulses, A.M. Weiner
• Aperture Modulated Diffusers (AMDs), H.P. Herzig and P. Kipfer
• Optical Properties of QuasiPeriodic Structures: Linear and Nonlinear Analysis, M. Bertolotti and C. Sibilia
• Diffractive Optical Elements in Materials Inspection, R. Silvennoinen, K.E. Peiponen and T. Asakura
• MultipleWavelength Interferometry for Absolute Distance Measurement, R. Dandliker and Y. Salvade
• Speckle Metrology—Some Newer Techniques and Applications, R.S. Sirohi
• Limits of Optical Range Sensors and How to Exploit Them, G. Hausler, P. Ettl, M. Schenk, G. Bohn and I. Laszlo
• Imaging Spectroscopy for the NonInvasive Investigations of Paintings, A. Casini, F. Lotti and M. Picollo
• Optical Coherence Tomography in Medicine, A.F. Fercher and C.K. Hitzenberger
• The Spectral Optimization of Human Vision: Some Paradoxes, Errors and Resolutions, B.H. Soffer and D.K. Lynch
• Optical Methods for Reproducing Sounds from Old Photograph Records, J. Uozumi and T. Asakura
The fifth book in the series appeared in August 2002 under the title International Trends in Applied Optics. It was edited by Past President of ICO, Arthur H. Guenther and published by SPIE. The book includes the following chapters:
• UltrashortPulse LaserMatter Interaction and Fast Instabilities, M. N. Libenson
xxv
ICO International Trends in Optics Series History
• Ultrafast Modelocked Lasers for the Measurement of Laser Frequencies and as Optical Clockworks, R. Holzwarth, T. Udem, and T. W. Hänsch
• Ablation of Metals with Femtosecond Laser Pulses, S. I. Anisimov
• Laser Microprocessing and Applications in Microelectronics and Electronics, Y. Feng Lu
• There are No Fundamental Limits to Optical Lithography, S. R. J. Brueck
Prototyping
• Laserproduced
Rapid
in
Manufacturing,
Y.
P.
Kathuria
• Computer Numerically Controlled Optics Fabrication, H. Pollicove and D. Golini
• Interference Coatings for the Ultraviolet Spectral Region, N. Kaiser
• Standardization in Optics Characterization, D. Ristau
• Advances in Thin Films, K. Lewis
• MicroOptics for Spectroscopy, R. Dändliker, H. P. Herzig, O. Manzardo, T. Scharf, and G. Boer
• Defense Optics and ElectroOptics , G. J. Simonis, G. Wood, Z. G. Sztankay, A. Goldberg, and J. Pellegrino
• Recent Progress in System and Component Technologies for Fiber Optic Communication, N. Shibata
• ShortDistance Optical Interconnections with VCSELs, H. Thienpont and V. Baukens
• Spontaneous Emission Manipulation, M. O. Scully, S. Zhu, and M. S. Zubairy
• Progress in Fiber Optics and Optical Telecommunication, A. K. Ghatak and B. P. Pal
• Nano and Atom Photonics, M. Ohtsu
• Binary Image Decompositions for Nonlinear Optical Correlations, H.H. Arsenault and P. GarcíaMartínez
• Optical Pattern Recognition of Partially Occluded Images, K. ChałasińskaMacukow
• Optical Sensing by Fiber and Integrated Optics Devices, G. C. Righini
• Waveoptical Engineering, F. Wyrowski and J. Turunen
• Neutron Optics, Neutron Waveguides, and Applications, M. L. Calvo and R. F. AlvarezEstrada
• Polarimetric Imaging, P. Réfrégier, F. Goudail, and P. Chavel
• Atmospheric Compensation, R. Q. Fugate
ICO International Trends in Optics Series History
xxvi
• Coherent Imaging Metrology in Life Sciences and Clinical Diagnostics, G. von Bally
• Optical Data Storage, H. Coufal and G. W. Burr
• Archaeological Optics, J. M. Enoch
• Thirty Years of Laser Applications in Conservation, R. Salimbeni, Roberto Pini, and S. Siano
Chapter 1
FirstOrder Optical Systems for Information Processing
Tatiana Alieva
Universidad Complutense de Madrid, Spain
1.1 Introduction
1.2 Canonical Integral Transforms: Deﬁnition and Classiﬁcation
1.2.1 Deﬁnition
1.2.2 Generalized imaging transforms
1.2.3 Orthosymplectic canonical transforms
1.2.4 Canonical transforms for the case det B = 0
1.3 Main Properties of the Canonical Integral Transforms
1.3.1 Parseval theorem
1.3.2 Shift theorem
1.3.3 Convolution theorem
1.3.4 Scaling theorem
1.3.5 Coordinates multiplication and derivation theorems
1.4 Canonical Integral Transforms of Selected Functions
1.4.1 Plane wave, chirp, and Gaussian functions
1.4.2 Periodic functions
1.4.3 Eigenfunctions for the canonical integral transforms
1.5 Generalized Convolution for Analog Optical Information Processing
1.5.1 Analog optical information processing
1.5.2 Generalized convolution: Deﬁnition
1.5.3 Filtering in fractional Fourier domains
1.5.4 Pattern recognition
1.5.5 Localization of the generalized chirp signals
1.5.6 Security applications
1.6 Other Optical Computing Approaches via Orthosymplectic Transforms
1.6.1
Mode presentation on orbital Poincaré sphere
2
Chapter 1
1.6.2 Orbital angular momentum manipulation
1.6.3 Geometric phase accumulation
References
1.1 Introduction
During the last decades, optics is playing an increasingly important role in acqui sition, processing, transmission, and archiving of information. In order to underline the contribution of optics in the information acquisition process, let us mention such optical modalities as microscopy, tomography, speckle imaging, spectroscopy, metrology, velocimetry, particle manipulation, etc. Data transmission through optical ﬁbers and optical data storage (CD, DVD, as well as current advances of holographic memories) make us everyday users of optical information technology. In the area of information processing, optics also has certain advantages with respect to electronic computing, thanks to its massive parallelism, operating with continuous data, possibility of direct penetration into the data acquisition process, implementation of fuzzy logic, etc. The basis of the analog coherent optical information processing is the ability of a thin convergent lens to perform the Fourier transform (FT). More than 40 years ago, Van der Lugt introduced an optical scheme for convolution/correlation operation, based on a cascade of two optical systems performing the Fourier transform with ﬁlter mask between them, initiating an era of Fourier optics. ^{1} This simple scheme realizes the most important shiftinvariant operations in signal/image processing, such as ﬁltering and pattern recognition. Nowadays, the Fourier optics area has been expanded with more sophisticated signal processing tools such as wavelets, bilinear distributions, fractional transformations, etc. Never theless, the paraxial optical systems (also called ﬁrstorder or Gaussian ones, which consist for example from several aligned lenses, or mirrors) remain the basic elements for analog optical information processing. In paraxial approximation of the scalar diffraction theory, a coherent light propagation through such a system is described by a canonical integral transform (CT). Thus starting from the complex ﬁeld amplitude at the input plane of the system, we have its CT at the output plane. The twodimensional CTs include, among others, such wellknown transformations as image rotation, scaling, fractional Fourier ^{2} and Fresnel transforms. We can say that the CTs represent a twodimensional signal in different phase space domains, where the phase space is deﬁned by the position and momentum (spatial frequency) coordinates. The signal manipulation in different phase space domains opens new perspectives for information processing. Indeed, several useful applications of the ﬁrstorder optical systems for information processing have been proposed in the past decade. In particular ﬁrstorder optical systems performing fractional Fourier transform have been used for shiftvariant ﬁltering, noise reduction, chirp localization, encryption, etc. ^{2}^{–}^{5} Others have served as mode converters, which transform the
FirstOrder Optical Systems for Information Processing
3
Hermite −Gaussian modes into helicoidal vortex Laguerre−Gaussian ones or other structurally stable modes. ^{6}^{,} ^{7} These modes, in particular, are interesting for new types of information encoding in orbital angular momentum of beam ^{8} or in the geometric phase accumulated when it undergoes the cyclic transformation. ^{9} Moreover the beam evolution in the ﬁrstorder optical systems is a good model for the analysis of twodimensional harmonic oscillator. ^{1}^{0} In this chapter, we brieﬂy summarize the main properties of the two dimensional CTs, ^{1}^{1} used for the description of the ﬁrstorder systems, consider their applications to traditional analog optical signal processing tasks, such as ﬁltering, pattern recognition, encryption, etc., and then discuss new methods of information encoding related to the orbital angular momentum transfer and geometric phase accumulation.
1.2 Canonical Integral Transforms: Deﬁnition and Classiﬁcation
1.2.1 Deﬁnition
The evolution of the complex ﬁeld amplitude f ( r ) during its propagation through a ﬁrstorder optical system is described by the linear integral transform
f _{o} ( r _{o} ) =
∞
−∞
f _{i} ( r _{i} ) K ^{t} (r _{i} , r _{o} ) d r _{i} ,
where subindices i and o stand for input and output planes of the system. The kernel K ^{t} ( r _{i} , r _{o} ) is parametrized by the wavelength λ and the real symplectic ray transformation 4 × 4 matrix t that relates the position r _{i} and direction q _{i} of an incoming ray to the position r _{o} and direction q _{o} of the outgoing ray,
^{} q
o
r o ^{} _{=} ^{} a
c
d r i b
q
i = t
r
q
i
i
^{.}
Proper normalization of the variables and the matrix parameters to some length
factor w and λ leads to the dimensionless variables: r = r / ^{√} λw , q = q ^{} w/λ, A = a , B = b/w , C =cw , D = d, which will be used further in this chapter,
r
q
o ^{} _{=} ^{} A
C
o
D B
i = T
r
q
i
i
r
q
i
,
(1.1)
where r = ( x, y ) ^{t} and q = ( q _{x} , q _{y} ) ^{t} . As usual, the superscript ^{t} denotes transpo sition. The normalized variable q can also be interpreted as spatial frequency or ray momentum. The canonical integral transform associated with matrix T will be represented by the operator R ^{T}
f _{o} ( r _{o} ) = R ^{T} [f _{i} ( r _{i} )] ( r _{o} ) = F _{T} ( r _{o} ) =
∞
−∞
f _{i} ( r _{i} ) K ^{T} (r _{i} , r _{o} ) d r _{i} .
(1.2)
4
Chapter 1
The CT is a linear transform: R ^{T} [f ( r _{i} ) + g ( r _{i} )] ( r ) = R ^{T} [f ( r _{i} )] ( r ) + R ^{T} [g ( r _{i} )] ( r ) . It is additive in the sense that R ^{T} ^{2} R ^{T} ^{1} = R ^{T} ^{2} ^{×} ^{T} ^{1} . The inverse transformation is parametrized by the matrix T ^{−} ^{1} , which, because T is symplectic, is given by
(1.3)
−C ^{t}
D ^{t}
_{T} − 1 _{=} ^{}
−B ^{t}
A ^{t}
.
Any proper normalized symplectic ray transformation matrix can be decom posed in the modiﬁed Iwasawa form as ^{1}^{2}
T = ^{A}
C
D = ^{B}
I
−G
0 ^{} ^{} S
0
I
S
− 1
0
X
−Y
Y _{X} = T _{L} T _{S} T _{O} ,
(1.4)
with I throughout denoting the identity matrix, in which the ﬁrst matrix represents
a lens transform described by the symmetric matrix
G = −( CA ^{t} + DB ^{t} )( AA ^{t} + BB ^{t} ) ^{−} ^{1} = G ^{t} .
(1.5)
The second matrix corresponds to a scaler described by the positive deﬁnite symmetric matrix
(1.6)
S = (AA ^{t} + BB ^{t} ) ^{1} ^{/} ^{2} = S ^{t}
and the third is an orthosymplectic ^{1}^{2}^{,} ^{1}^{3} (i.e., both orthogonal and symplectic) matrix, which can be shortly represented by the unitary matrix
U = X + i Y = (AA ^{t} + BB ^{t} ) ^{−} ^{1} ^{/} ^{2} ( A + i B ) .
(1.7)
Note that because A = SX and B = SY , the products B ^{−} ^{1} A = Y ^{−} ^{1} X and A ^{−} ^{1} B = X ^{−} ^{1} Y used further in different relations are deﬁned by the orthogonal matrix T _{O} . Because the ray transformation matrix T is symplectic and therefore
AB ^{t} = BA ^{t} , CD ^{t} = DC ^{t} , AD ^{t} − BC ^{t} = I ,
(1.8)
A ^{t} C = C ^{t} A , B ^{t} D = D ^{t} B , A ^{t} D − C ^{t} B = I ,
it has only ten free parameters. We call the transform associated with T separable
if the block matrices A , B , C, and D and G , S, X, and Y correspondingly are diagonal. A separable transform has six degrees of freedom which reduce to three for rotational symmetric case corresponding to scalar block matrices.
In the oftenused case det B = 0, the CT takes the form of Collins’ integral ^{1}^{4}
f _{o} ( r _{o} ) = R ^{T} [ f _{i} ( r _{i} )] ( r _{o} ) = (det i B ) ^{−} ^{1} ^{/} ^{2}
∞
−∞
f _{i} ( r _{i} )
× exp ^{} i π ^{} r ^{t} B ^{−} ^{1} Ar _{i} − 2 r ^{t} B ^{−} ^{1} r _{o} + r _{o} ^{t} DB ^{−} ^{1} r _{o} ^{}^{} d r _{i} . (1.9)
i
i
The kernel corresponds to twodimensional generalized chirp function because its phase is a polynomial of second degree of variables r _{i} and r _{o} . In particular
FirstOrder Optical Systems for Information Processing
5
for A = 0, the kernel as a function of r _{i} has a form of plane wave. Thus for
A = D = 0 and B = − C = I , we obtain apart from a constant phase factor
exp(−i π / 2) the Fourier transform F [f ( r _{i} )](r _{o} )
F [f ( r _{i} )](r _{o} ) =
∞
−∞
f ( r _{i} ) exp(−i 2 π r _{o} ^{t} r _{i} ) d r _{i} ,
(1.10)
known in optics as an angular spectrum of the complex ﬁeld amplitude f . Moreover from the analysis of the matrix Q = B ^{−} ^{1} A = Y ^{−} ^{1} X , it follows that the kernel as a function of r _{i} corresponds to the elliptic, hyperbolic or parabolic waves if Q = 4Q _{1}_{1} Q _{2}_{2} − ( Q _{1}_{2} + Q _{2}_{1} ) ^{2} is positive, negative, or 0, relatively. The wellknown Fresnel transform that describes the evolution of the complex ﬁeld amplitude during light propagation in an isotropic homogeneous medium
at distance z , which in this chapter is a normalized dimensionless variable, is
associated with matrix T _{F} ( z ) : A = D = I , C = 0 , B =z I .
1.2.2 Generalized imaging transforms
The case B = 0 corresponds to the generalized imaging condition
f _{o} ( r )=( det A ) ^{−} ^{1} ^{/}^{2} exp ^{} i π r ^{t} CA ^{−} ^{1} r ^{} f _{i} ( A ^{−} ^{1} r ) ,
(1.11)
which includes a possible scaling and rotation of the input function accompanied by an additional phase modulation. In the case C = B = 0 , we have a family of the imaging transforms without phase modulation, which includes image rotation, scaling, and shearing. The rotator transform, associated with T _{r} (α ) : C _{r} = B _{r} = 0 and
A _{r} = D _{r} = X _{r} =
− sin α cos α ^{,}
cos α
sin α
produces a clockwise rotation of f _{i} in x − y plane and, correspondingly, its FT (the angular spectrum) in q _{x} − q _{y} plane at angle α
f _{o} ( x, y ) =
f _{i} ( x cos α − y sin α , x sin α + y cos α ) .
A ﬂexible optical scheme performing a rotation at angle α by only the appropriate
rotating of cylindrical lenses composing the setup has been recently proposed. ^{1}^{5} Alternatively, Dove prisms can be used for optical rotator realization. The separable scaling transform associated with the block matrices
A = D ^{−} ^{1} = S _{s} , C = B = 0, where
S s = ^{s} 0 ^{x}
^{0}
s
y ^{,}
6
Chapter 1
leads to f _{o} ( x, y )=(s _{x} s _{y} ) ^{−} ^{1} ^{/}^{2} f _{i} (x/s _{x} , y/s _{y} ). If we combine the separable scaling together with pre and postrotations, then the afﬁne imaging transfor mation is obtained. It is parametrized by the matrix T _{r} (β) T _{S} _{s} ( s _{x} , s _{y} ) T _{r} ( α ) with A _{a}_{i} = ( D _{a}_{i} ^{t} ) ^{−} ^{1} , C _{a}_{i} = B _{a}_{i} = 0 , where
A ai
=
−s _{x} cos α sin β − s _{y} sin α cos β −s _{x} sin α sin β + s _{y} cos α cos β
s _{x} cos α cos β − s _{y} sin α sin β
s _{x} sin α cos β + s _{y} cos α sin β
_{=} ^{} s xx s xy s
yx
s
yy
,
(1.12)
and leads to the following function transformation:
f _{o} ( x, y )=( s _{x}_{x} s _{y}_{y} − s _{x}_{y} s _{y}_{x} ) ^{−} ^{1} ^{/}^{2} f _{i}
s yy x − s xy y
_{,}
s xx s yy − s xy s yx
s xx y − s yx x
yx ^{.}
s xx s yy − s xy s
Note that for α = −β the matrix A _{a}_{i} = A _{a}_{i} ^{t} is symmetric and reduces to the scaler deﬁned in Eq. (1.4). The shearing operation in (x, y ) and, correspondingly, in (q _{x} , q _{y} ) planes, which is a particular case of the afﬁne image transformer, is described by A _{a}_{i} with diagonal elements equal to 1 and one antidiagonal element equals to 0 . For
example, for β = α + π/ 2 , u = −2 cot 2α , and s _{x} = s operation f _{o} ( x, y ) = f _{i} ( x − uy, y ) , associated with the matrices
− tan α , the skew
−
y
1
=
A sh =
1
0
u
1
,
D sh =
1
−u
0
1
,
B _{s}_{h} = C _{s}_{h} = 0,
(1.13)
is performed. Note that the Fresnel and spherical lens transformations correspond to shearing in the ( x, q _{x} ) and ( y, q _{y} ) planes of the phase space. ^{1}^{7} The ﬁrst of them does not belong to the imaging type transforms. The lens transform described by T _{L} , Eq. (1.4), with the block matrix
_{G} _{=} ^{} g g xx
xy
g
g
xy
yy
(1.14)
produces the phase modulation of the input wavefront with polynomial of second degree
f _{o} ( x, y ) = exp ^{} −i π ^{} g _{x}_{x} x ^{2} + 2g _{x}_{y} xy + g _{y}_{y} y ^{2} ^{}^{} f _{i} ( x, y ) .
In practice, the generalized lens ^{1}^{6} corresponds to the combination of n aligned cylindrical lenses of power p _{j} ( p _{j} > 0 for convergent lens), which are attached one to another and counterclockwise rotated with respect to the transversal OX axis at angles φ _{j} . Then g _{x}_{x} = ^{} _{=}_{1} p _{j} cos ^{2} φ _{j} , g _{x}_{y} = − ^{} _{=}_{1} p _{j} (sin 2φ _{j} ) / 2 , and g _{y}_{y} = ^{} _{=}_{1} p _{j} sin ^{2} φ _{j} . Depending on the angles and the focal distances of
n
j
n
j
n
j
FirstOrder Optical Systems for Information Processing
7
the cylindrical lenses, we obtain the elliptic (including spherical), hyperbolic, or parabolic phase modulations. Note that the generalized lens transform can be represented as a separable one embedded into direct, inverse rotators, which can be written in the matrix form
g _{y} cos ^{2} α , g _{y}_{y} =
as
T _{L} = T _{r} ( − α ) T _{L} ( g _{x} , g _{y} ) T _{r} ( α ) , where g _{x}_{x} = g _{x} cos ^{2} α + g _{x} sin ^{2} α + g _{y} cos ^{2} α , and g _{x}_{y} = (g _{x} − g _{y} ) cos α sin α .
s
1.2.3 Orthosymplectic canonical transforms
Because the rotator matrix T _{r} ( α ) is orthogonal, then the rotator belongs not only to the imaging transform class but also to the orthosymplectic one (CTs associated with orthogonal ray transformation matrix). Thus, it can be expressed by the unitary
matrix (1.7) U r ( α )= _{}
(1.15)
cos α − sin α cos
sin α
_{α} .
There are two other basic orthosymplectic CTs, which also have the rotational character: the separable fractional FT and the gyrator transform. The separable fractional FT [ray transformation matrix T _{f} ( γ _{x} , γ _{y} ) ] is described by the unitary matrix
U _{f} ( γ _{x} , γ _{y} )= ^{e}^{x}^{p}^{(}^{i} ^{γ} ^{x} ^{)}
0
_{e}_{x}_{p}_{(}_{i} _{γ} _{y} _{)} ^{,}
^{0}
^{(}^{1}^{.}^{1}^{6}^{)}
which corresponds to rotations in the (x, q _{x} ) and ( y, q _{y} ) planes through the angles γ _{x} and γ _{y} , respectively. The kernel of the fractional FT is a product of two similar
ones, K ^{T} f ^{(} ^{γ} ^{x} ^{,} ^{γ} ^{y} ^{)} ( r _{i} , r _{o} ) = K
f
x
(
x _{i} , x _{o} ) K
γ
f
y
(
y _{i} , y _{o} ) , with K
γ
γ
f
x
(
x _{i} , x _{o} ) given by
K
γ
f
x
(x _{i} , x _{o} )=( i sin γ _{x} ) ^{−} ^{1} ^{/}^{2} exp i π ^{(} ^{x}
2
i
+ x _{o} ) cos γ _{x} − 2 x _{o} x _{i} sin γ _{x}
2
.
(1.17)
If γ _{x} = γ _{y} = ϕ , then the fractional FT is symmetric. For ϕ = 0, it corresponds to the identity transform K ( x _{i} , x _{o} ) = δ ( x _{i} − x _{o} ) and, for ϕ = π/ 2 , to the common Fourier transform (1.10) apart from constant −i . Two simple optical schemes performing the symmetric fractional FT have been proposed. ^{1}^{7} The ﬁrst one consists of a thin spherical convergent lens of power p located at the equal distances z = 2p ^{−} ^{1} sin ^{2} ( ϕ / 2) between the input and output planes. Then the output complex ﬁeld amplitude is the symmetric fractional FT at angle ϕ of the input one. The second setup consists of two identical spherical convergent lenses of power p located at the input and output system plane with the distance z = 2p ^{−} ^{1} sin ^{2} ( ϕ / 2) between them. Moreover, the propagation of the optical beam through the optical ﬁber with a quadratic refractive index proﬁle also produces the symmetric fractional FT at angles deﬁned by the propagation distance and the refractive index gradient. ^{1}^{8}^{,} ^{1}^{9} If γ _{x} = −γ _{y} = γ , then we have the antisymmetric fractional FT. The combi nation of the symmetric R ^{T} f ^{(}^{ϕ} ^{,} ^{ϕ} ^{)} and antisymmetric R ^{T} f ^{(} ^{γ}^{,} ^{−} ^{γ}^{)} fractional FTs
0
f
8
Chapter 1
deﬁnes the separable fractional FT R ^{T} f ^{(}^{γ} ^{x} ^{,} ^{γ} ^{y} ^{)} at angles γ _{x} = ϕ + γ and γ _{y} = ϕ − γ . More information about the fractional FT can be found in Refs. 2, 3, 17–19 and references there in. The gyrator transform (GT), associated with T _{g} ( ϑ ) , corresponds to twisting, i.e., rotations in the ( x, q _{y} ) and (y, q _{x} ) planes of phase space, and is described by unitary matrix
(1.18)
U _{g} ( ϑ )=
cos ϑ i sin ϑ
i sin ϑ cos ϑ
.
The kernel of the GT has a form of the hyperbolic wave
K ^{T} ^{g} ^{(} ^{ϑ}^{)} (r _{i} , r _{o} ) =
sin ϑ  _{e}_{x}_{p} ^{} _{i} _{2} _{π} ( x _{o} y _{o} + x _{i} y _{i} ) cos ϑ − (x _{i} y _{o} + x _{o} y _{i} )
1
sin ϑ
,
(1.19)
which reduces to δ ( r _{i} − r _{o} ) for ϑ = 0, to δ ( r _{i} + r _{o} ) for ϑ = π , and to the twisted FT kernel exp [∓i 2 π ( x _{i} y _{o} + x _{o} y _{i} )] for ϑ = ±π / 2 . A detailed analysis of the GT can be found in Refs. 20–22. Based on the matrix formalism ﬂexible optical setups, which perform the antisymmetric fractional FT R ^{T} f ^{(}^{γ}^{,} ^{−} ^{γ}^{)} , and the GT R ^{T} ^{g} ^{(} ^{ϑ}^{)} have been designed. ^{1}^{5}^{,} ^{2}^{1} These optical schemes contain three generalized lenses, L _{1} , L _{2} , and L _{3} = L _{1} , with ﬁxed equal distances between them denoted by z . The trans formation angle is changed by rotation of the cylindrical lenses which form the generalized lenses. In the case of the antisymmetric fractional FT setup every generalized lens L _{j}
( j = 1, 2 ) consists from a convergent spherical lens with focal distance p
and convergent and divergent cylindrical lenses with focal distances ±z rotated at
= π/ 2 − φ ^{(}^{j} ^{)} with respect to OX axis. The antisymmetric
angle φ
= z/j
−
j
1
(
j )
1
= φ ^{(} ^{j} ^{)} , φ
2
(
j )
fractional FT at angle ( γ , −γ) is achieved if cos(2 φ ^{(}^{1}^{)} ) = cot( γ / 2) and 2 φ ^{(}^{2}^{)} = π/ 2 − γ. It is easy to see from the last relation that this setup is able to perform the antisymmetric fractional FT for the angles [ π / 2 , 3 π / 2] that cover a π interval needed for the different applications.
In the case of the GT, every generalized lens L _{j} ( j = 1, 2 ) is a combination of two convergent cylindrical lenses of equal focus distance z/j rotated at angle
φ
= φ ^{(} ^{j} ^{)} − π /2 . The GT at angle ϑ is achieved if cos(2 φ ^{(}^{1}^{)} ) =
cot( ϑ /2) and sin(2φ ^{(}^{2}^{)} ) = (sin ϑ ) /2. We again observe that this setup is able to perform the GT for the angles from π interval [π / 2 , 3 π / 2] . It has been shown ^{2}^{3} that any orthosymplectic matrix can be decomposed in the form
T _{O} = T _{r} ( β ) T _{f} ( γ _{x} , γ _{y} ) T _{r} ( α ) .
(1.20)
(
j )
1
= −φ ^{(} ^{j} ^{)} , φ
2
(
j )
It means that R ^{T} O is a separable fractional Fourier transformer R ^{T} f embedded between two rotators R ^{T} ^{r} . In particular for the gyrator matrix, we obtain T _{g} ( ϑ ) = T _{r} ( − π / 4) T _{f} ( ϑ , −ϑ ) T _{r} ( π /4). Therefore, based on the optical setups,
FirstOrder Optical Systems for Information Processing
9
performing the fractional FT and rotator a system for arbitrary orthosymplectic transformation can be constructed. Moreover, because the generalized lens transform and scaler from the decom position (1.4) also can be presented as separable ones embedded into two rotators, any CT can be written in the form
T _{r} ( α _{4} ) T _{L} _{s} ( g _{x} , g _{y} ) T _{r} ( α _{3} ) T _{S} _{s} ( s _{x} , s _{y} ) T _{r} ( α _{2} ) T _{f} ( γ _{x} , γ _{y} ) T _{r} ( α _{1} ) , (1.21)
T =
where we used the angle additivity of the rotator transform. The 10 parameters of
4 ) and 2 parameters
for three separable transforms: lens (g _{x} , g _{y} ) , scaler ( s _{x} , s _{y} ) , and fractional FT
( γ _{x} , γ _{y} ) . In general, the ray matrix decomposition into a cascade of the others is very useful for the analysis of a particular optical system. Thus, the Fresnel transform is a combination of the symmetric fractional FT at angle arctan z , scaling with
s = ^{√} 1 + z ^{2} and spherical lens transform with g = z/ (1 + z ^{2} ) . On the other side, the Fresnel transform itself can be considered as a basic element for the system design, because the abovementioned fractional Fourier and gyrator trans formers are constructed as a cascade of lenses and homogeneous medium intervals, described by the Fresnel transform.
the CT are deﬁned then as 4 rotation angles α _{j} ( j = 1,
,
1.2.4 Canonical transforms for the case det B = 0
In order to deal with the singular case det B = 0 , but B = 0, we can use the Iwasawa decomposition (1.4), from which we note that B = SY . Because the scaling matrix S is nonsingular, a singularity of B is only due to the orthosym plectic part described by T _{O} . Moreover from the analysis of Eq. (1.20), we conclude that the separable fractional Fourier transformer is responsible for a singularity of the submatrix B . Thus, det B = 0 if for at least one coordinate the fractional Fourier transformer acts as an identity system, i.e., γ _{x} = 0 and/or γ _{y} = 0. Then, based on the modiﬁed Iwasawa decomposition (1.4) and Eq. (1.20), we can write a general representation of the CT, which is valid for any ray transfor mation matrix, including a singular submatrix B ^{2}^{3}
f _{o} ( r _{o} ) = R ^{T} [ f _{i} ( r _{i} )] ( r _{o} ) = (det S) ^{−} ^{1} ^{/} ^{2} exp(−i π r _{o} ^{t} Gr _{o} )
×R
Гораздо больше, чем просто документы.
Откройте для себя все, что может предложить Scribd, включая книги и аудиокниги от крупных издательств.
Отменить можно в любой момент.