Steven M. Kay - Fundamentals of Statistical Signal Processing, Volume III: Practical Algorithm Development

Practical Algorithm Development

Steven M. Kay

Upper Saddle River, NJ Boston Indianapolis San Francisco
New York Toronto Montreal London Munich Paris Madrid
Capetown Sydney Tokyo Singapore Mexico City

Many of the designations used by manufacturers and sellers to distinguish their products are claimed as trademarks. Where those designations appear in this book, and the publisher was aware of a trademark claim, the designations have been printed with initial capital letters or in all capitals.

The author and publisher have taken care in the preparation of this book, but make no expressed or implied warranty of any kind and assume no responsibility for errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of the use of the information or programs contained herein.

The publisher offers excellent discounts on this book when ordered in quantity for bulk purchases or special sales, which may include electronic versions and/or custom covers and content particular to your business, training goals, marketing focus, and branding interests. For more information, please contact:

U.S. Corporate and Government Sales
(800) 382-3419

For sales outside the United States please contact:

International Sales

Visit us on the Web: informit.com/ph

Library of Congress Control Number: 92029495

Copyright 2013 Pearson Education, Inc.

All rights reserved. Printed in the United States of America. This publication is protected by copyright, and permission must be obtained from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying, recording, or likewise. To obtain permission to use material from this work, please submit a written request to Pearson Education, Inc., Permissions Department, One Lake Street, Upper Saddle River, New Jersey 07458, or you may fax your request to (201) 236-3290.

ISBN-13: 978-0-13-280803-3
ISBN-10: 0-13-280803-X

Text printed in the United States on recycled paper at Courier Corporation in Westford, Massachusetts
First printing, March 2013.

To my wife Cindy, who is and will always be my anchor

Fundamentals of Statistical Signal Processing: Practical Algorithm Development is the third volume in a series of textbooks by the same name. Previous volumes described the underlying theory of estimation and detection algorithms. In contrast, the current volume addresses the practice of converting this theory into software algorithms that may be implemented on a digital computer. In describing the methodology and techniques, it will not be assumed that the reader has studied the first two volumes, but of course, he/she is certainly encouraged to do so. Instead, the descriptions will focus on the general concepts using a minimum of mathematics but will be amply illustrated using MATLAB implementations. It is envisioned that the current book will appeal to engineers and scientists in industry and academia who would like to solve statistical signal processing problems through design of well-performing and implementable algorithms for real systems. These systems are typically encountered in many signal processing disciplines, including but not limited to communications, radar, sonar, biomedical, speech, optical, and image processing. Additionally, due to the emphasis on actual working algorithms, the material should be of use to the myriad of researchers in statistical signal processing who wish to obtain an overview of the state of the practical art. Those new to the field who are concerned with sorting the wheat from the chaff in the ever-exploding arsenal of signal processing algorithms will also benefit from the exposition.

The overall goal for this book is to allow the reader to develop his/her intuition and subsequent expertise into the practice of statistical signal processing. To accomplish this goal we have endeavored to

Describe the methodology, including mathematical modeling, computer simulation, and performance evaluation, used to develop algorithms.

Allow the reader to assimilate the important concepts by practicing with the tools typically available. These include useful analytical results and MATLAB implementations for design, evaluation, and testing.

Highlight the approaches and specific algorithms that work in practice, i.e., those that have stood the test of time.

Illustrate application areas by describing and solving real-world problems.

Introduce the reader to some extensions required in practice.

Translate a mathematical algorithm into MATLAB code and verify the integrity of the solution.

Pedagogically speaking, we believe that the strong reliance on MATLAB examples will aid in understanding the workings and subtleties of the various algorithms. The reader will then learn by doing. In the same vein, numerous analytical exercises have been embedded into the text for student practice. The full solutions are contained in the appendices of each chapter. MATLAB exercises are also given, with abbreviated solutions listed in the appendix of each chapter, and the full solutions, including executable MATLAB code, contained on the enclosed CD. At the end of many of the chapters is a section called Lessons Learned. These conclusions are important observations that are intended to provide insight into the inner workings of the algorithms and rules of thumb that are routinely employed. These lessons learned are often critical to the development of successful algorithms. Most of the topics chosen for inclusion have been drawn from Fundamentals of Statistical Signal Processing: Estimation Theory, 1993, and Fundamentals of Statistical Signal Processing: Detection Theory, 1998, but we have also added much material from Modern Spectral Estimation: Theory and Application, 1988 (all books published by Prentice Hall), since the latter book contains many of the techniques required for data simulation and analysis. Finally, it is hoped that the current book will be useful for self-study. Although this volume can be used without MATLAB as a practice tool, much of the understanding that comes from that experience would be lost.

The background assumed for the reader is a knowledge of calculus, basic linear systems, including some digital signal processing, probability and introductory random processes, and some linear and matrix algebra. As previously mentioned, we have attempted to describe the techniques without heavy reliance upon mathematics and this background material. However, in the end the algorithms are by their nature mathematical and so it must be that this goal can only partially be attained.

The author would like to acknowledge the contributions of the many people who over the years have provided stimulating discussions of teaching and research problems and opportunities to apply the results of that research. Thanks are due to my colleagues L. Jackson, R. Kumaresan, L. Pakula, and P. Swaszek of the University of Rhode Island. A debt of gratitude is owed to all my current and former graduate students. They have contributed to the final manuscript through many hours of pedagogical and research discussions as well as by their specific comments and questions. In particular, Quan Ding and Naresh Vankayalapati have contributed specific comments and helped with the exercise solutions. Additionally, William Knight has provided valuable feedback on the manuscript. The author is indebted to the many agencies and program managers who have sponsored his research. These managers include Jon Davis, Darren Emge, James Kelly, Muralidhar Rangaswamy, Jon Sjogren, and Peter Zulch. The agencies include the Naval Undersea Warfare .