100% found this document useful (7 votes)

67 views

Download full Fundamentals of Digital Signal Processing Using MATLAB 2nd Edition Robert J. Schilling ebook all chapters

Robert

Uploaded by

zellijerve

Available Formats

Download as PDF, TXT or read online on Scribd

100% found this document useful (7 votes)

67 views

Download full Fundamentals of Digital Signal Processing Using MATLAB 2nd Edition Robert J. Schilling ebook all chapters

Robert

Uploaded by

zellijerve

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 81

Download the full version of the ebook at

https://ebookultra.com

Fundamentals of Digital Signal Processing

Using MATLAB 2nd Edition Robert J. Schilling

https://ebookultra.com/download/fundamentals-of-
digital-signal-processing-using-matlab-2nd-
edition-robert-j-schilling/

Explore and download more ebook at https://ebookultra.com

Recommended digital products (PDF, EPUB, MOBI) that
you can download immediately if you are interested.

Digital signal processing with examples in MATLAB Second

Edition Hush

https://ebookultra.com/download/digital-signal-processing-with-
examples-in-matlab-second-edition-hush/

ebookultra.com

Digital Signal and Image Processing using MATLAB Volume 2

Advances and Applications The Deterministic Case 2nd
Edition Gérard Blanchet
https://ebookultra.com/download/digital-signal-and-image-processing-
using-matlab-volume-2-advances-and-applications-the-deterministic-
case-2nd-edition-gerard-blanchet/
ebookultra.com

Signal Processing for Digital Communications Artech House

Signal Processing Library George J. Miao

https://ebookultra.com/download/signal-processing-for-digital-
communications-artech-house-signal-processing-library-george-j-miao/

ebookultra.com

Digital Signal Processing Second Edition Fundamentals and

Applications Li Tan

https://ebookultra.com/download/digital-signal-processing-second-
edition-fundamentals-and-applications-li-tan/

ebookultra.com
Digital Signal Processing and Applications with the
TMS320C6713 and TMS320C6416 DSK Topics in Digital Signal
Processing 2nd Edition Rulph Chassaing
https://ebookultra.com/download/digital-signal-processing-and-
applications-with-the-tms320c6713-and-tms320c6416-dsk-topics-in-
digital-signal-processing-2nd-edition-rulph-chassaing/
ebookultra.com

Digital Signal Processing Laboratory 2nd ed Edition B.

Preetham Kumar

https://ebookultra.com/download/digital-signal-processing-
laboratory-2nd-ed-edition-b-preetham-kumar/

ebookultra.com

Digital Audio Signal Processing 2nd ed Edition Udo Zölzer

https://ebookultra.com/download/digital-audio-signal-processing-2nd-
ed-edition-udo-zolzer/

ebookultra.com

Digital Signal Processing Laboratory Second Edition Kumar

https://ebookultra.com/download/digital-signal-processing-laboratory-
second-edition-kumar/

ebookultra.com

Hack Audio An Introduction to Computer Programming and

Digital Signal Processing in MATLAB 2019th Edition Eric
Tarr
https://ebookultra.com/download/hack-audio-an-introduction-to-
computer-programming-and-digital-signal-processing-in-matlab-2019th-
edition-eric-tarr/
ebookultra.com
Fundamentals of Digital Signal Processing Using
MATLAB 2nd Edition Robert J. Schilling Digital Instant
Download
Author(s): Robert J. Schilling, Sandra L Harris
ISBN(s): 9780840069092, 084006909X
Edition: 2
File Details: PDF, 20.09 MB
Year: 2011
Language: english
SCHILLING
HARRIS

DIGITAL SIGNAL PROCESSING

FUNDAMENTALS OF
MATLAB®
using
SECOND
EDITION

To learn more about Cengage Learning, visit www.cengage.com

For your course and learning solutions, visit www.cengage.com/engineering
Purchase any of our products at your local bookstore or at our preferred
online store www.cengagebrain.com

9780840069092_cvr_use_wkg.indd 1 09/11/10 4:34 PM

Schilling-1120949 6909X˙00˙FM˙pi-xviii November 12, 2010 8:25
Schilling-1120949 6909X˙00˙FM˙pi-xviii November 12, 2010 8:25

Fundamentals of Digital
Signal Processing
Using MATLAB®
Second Edition

Robert J. Schilling and Sandra L. Harris

Clarkson University
Potsdam, NY

Australia • Brazil • Japan • Korea • Mexico • Singapore • Spain • United Kingdom • United States

i
This is an electronic version of the print textbook. Due to electronic rights
restrictions, some third party content may be suppressed. Editorial
review has deemed that any suppressed content does not materially
affect the overall learning experience. The publisher reserves the
right to remove content from this title at any time if subsequent
rights restrictions require it. For valuable information on pricing, previous
editions, changes to current editions, and alternate formats, please visit
www.cengage.com/highered to search by ISBN#, author, title, or keyword
for materials in your areas of interest.

Fundamentals of Digital Signal © 2012, 2005 Cengage Learning

Processing Using MATLAB®
ALL RIGHTS RESERVED. No part of this work covered by the copyright
Robert J. Schilling and Sandra L. Harris herein may be reproduced, transmitted, stored, or used in any form or
by any means graphic, electronic, or mechanical, including but not
Publisher, Global Engineering:
limited to photocopying, recording, scanning, digitizing, taping, web
Christopher M. Shortt
distribution, information networks, or information storage and
Acquisitions Editor: Swati Meherishi retrieval systems, except as permitted under Section 107 or 108 of
the 1976 United States Copyright Act, without the prior written
Senior Developmental Editor: Hilda Gowans
permission of the publisher.
Editorial Assistant: Tanya Altieri
Team Assistant: Carly Rizzo For product information and technology assistance, contact us at
Marketing Manager: Lauren Betsos Cengage Learning Customer & Sales Support, 1-800-354-9706.
For permission to use material from this text or product,
Media Editor: Chris Valentine
submit all requests online at www.cengage.com/permissions.
Content Project Manager: Further permissions questions can be e-mailed to
D. Jean Buttrom [email protected].
Production Service:
RPK Editorial Services, Inc.
Library of Congress Control Number: 2010938463
Copyeditor: Shelly Gerger-Knechtl
ISBN-13: 978-0-8400-6909-2
Proofreader: Becky Taylor
ISBN-10: 0-8400-6909-X
Indexer: Shelly Gerger-Knechtl
Compositor: MPS Limited, Cengage Learning
a Macmillan Company 200 First Stamford Place, Suite 400
Stamford, CT06902
Senior Art Director: Michelle Kunkler
USA
Internal Designer: Carmela Periera
Cengage Learning is a leading provider of customized learning solutions
Cover Designer: Andrew Adams
with ofﬁce locations around the globe, including Singapore, the United
Cover Image: © prudkov/Shutterstock Kingdom, Australia, Mexico, Brazil, and Japan. Locate your local ofﬁce at:
Rights Acquisitions Specialist: John Hill international.cengage.com/region.

Text and Image Permissions Researcher:

Cengage Learning products are represented in Canada by
Kristiina Paul
Nelson Education, Ltd.
First Print Buyer: Arethea L. Thomas
For your course and learning solutions, visit
www.cengage.com/engineering.

Purchase any of our products at your local college store or at our

preferred online store www.cengagebrain.com.

MATLAB is a registered trademark of The MathWorks,

3 Apple Hill Drive, Natick, MA 01760.

Printed in the United States of America

1 2 3 4 5 6 7 14 13 12 11 10

ii
Schilling-1120949 6909X˙00˙FM˙pi-xviii November 12, 2010 8:25

In memory of our fathers:

Edgar J. Schilling
and
George W. Harris

iii
Schilling-1120949 6909X˙00˙FM˙pi-xviii November 12, 2010 8:25

Preface

Digital signal processing, more commonly known as DSP, is a field of study with increasingly
widespread applications in the modern technological world. This book focuses on the fun-
damentals of digital signal processing with an emphasis on practical applications. The text,
Fundamentals of Digital Signal Processing, consists of the three parts pictured in Figure 1.

FIGURE 1: Parts
of Text
I. Signal and System Analysis

1. Signal Processing
2. Discrete-time Systems in the Time Domain
3. Discrete-time Systems in the Frequency Domain
4. Fourier Transforms and Signal Spectra

II. Digital Filter Design

5. Filter Design Specifications

6. FIR Filter Design
7. IIR Filter Design

III. Advanced Signal Processing

8. Multirate Signal Processing

9. Adaptive Signal Processing

vi Preface

• • • • • • • • • ••••• ••

Audience and Prerequisites

This book is targeted primarily toward second-semester juniors, seniors, and beginning gradu-
ate students in electrical and computer engineering and related fields that rely on digital signal
processing. It is assumed that the students have taken a circuits course, or a signals and systems
course, or a mathematics course that includes an introduction to the Fourier transform and the
Laplace transform. There is enough material, and sufficient flexibility in the way it can be
covered, to provide for courses of different lengths without adding supplementary material.
Exposure to MATLAB® programming is useful, but it is not essential. Graphical user interface
(GUI) modules are included at the end of each chapter that allow students to interactively
explore signal processing concepts and techniques without any need for programming. MAT-
LAB computation problems are supplied for those users who are familiar with MATLAB, and
are interested in developing their own programs.
This book is written in an informal style that endeavors to provide motivation for each
new topic, and features a careful transition between topics. Significant terms are set apart
for convenient reference using Margin Notes and Definitions. Important results are stated as
Propositions in order to highlight their significance, and Algorithms are included to summarize
the steps used to implement important design procedures. In order to motivate students with
examples that are of direct interest, many of the examples feature the processing of speech and
music. This theme is also a focus of the course software that includes a facility for recording and
playing back speech and sound on a standard PC. This way, students can experience directly
the effects of various signal processing techniques.

• • • • • • • • • ••••• ••

Chapter Structure
Each of the chapters of this book follows the template shown in Figure 2. Chapters start with
motivation sections that introduce one or more examples of practical problems that can be
solved using techniques covered in the chapter. The main body of each chapter is used to

FIGURE 2: Chapter
Structure
Motivation

Concepts,
techniques,
examples

GUI software,
case studies

Problems

Preface vii

introduce a series of analysis tools and signal processing techniques. Within these sections,
the analysis methods and processing techniques evolve from the simple to the more complex.
Sections marked with a ∗ near the end of the chapter denote more advanced or specialized
material that can be skipped without loss of continuity. Numerous examples are used throughout
to illustrate the principles involved.
Near the end of each chapter is a GUI software and case studies section that introduces
GUI modules designed to allow the student to interactively explore the chapter concepts and
techniques without any need for programming. The GUI modules feature a standard user
interface that is simple to use and easy to learn. Data files created as output from one module
can be imported as input into other modules. This section also includes case study examples
that present complete solutions to practical problems in the form of MATLAB programs.
The Chapter Summary section concisely reviews important concepts, and it provides a list of
student learning outcomes for each section. The chapter concludes with an extensive set of
homework problems separated into three categories and cross referenced to the sections. The
Analysis and Design problems can be done by hand or with a calculator. They are used to test
student understanding of, and in some cases extend, the chapter material. The GUI Simulation
problems allow the student to interactively explore processing and design techniques using the
chapter GUI modules. No programming is required for these problems. MATLAB Computation
problems are provided that require the user to write programs that apply the signal processing
√
techniques covered in the chapter. Solutions to selected problems, marked with the symbol,
are available as pdf files using the course software.

• • • • • • • • • ••••• ••

FDSP Toolbox
One of the unique features of this textbook is an integrated software package called the Fun-
damentals of Digital Signal Processing (FDSP) Toolbox that can be downloaded from the
companion web site of the publisher. It is also possible to download the FDSP toolbox from
the following web site maintained by the authors. Questions and comments concerning the
text and the software can be addressed to the authors at: [email protected].
www.clarkson.edu/~rschilli/fdsp

The FDSP toolbox includes the chapter GUI modules, a library of signal processing functions,
all of the MATLAB examples, figures, and tables that appear in the text, solutions to selected
problems, and on-line help . All of the course software can be accessed easily through a simple
menu-based FDSP driver program that is executed with the following command from the
MATLAB command prompt.
>> f_dsp

The FDSP toolbox is self-contained in the sense that only the standard MATLAB interpreter
is required. There is no need to for users to have access to optional MATLAB toolboxes such
as the Signal Processing and Filter Design toolboxes.

• • • • • • • • • ••••• ••

Support Material
To access additional course materials [including CourseMate], please visit www.cengagebrain
.com. At the cengagebrain.com home page, search for the ISBN of your title (from the back

viii Preface

cover of your book) using the search box at the top of the page. This will take you to the
product page where these resources can be found.
Supplementary course material is provided for both the student and√the instructor. For the
student, solutions to selected end-of-chapter problems, marked with a , are included as pdf
files with the FDSP toolbox. Students are encouraged to use these problems as a test of their
understanding of the material. For the instructor, an enhanced version of the FDSP toolbox
includes pdf file solutions to all of the problems that appear at the end of each chapter. In
addition, as an instructional aid, every computational example, every figure, every table, and
the solution to every problem in the text can be displayed in the classroom using the instructor’s
version of the driver module, f dsp.

• • • • • • • • • ••••• ••

Acknowledgments
This project has been years in the making and many individuals have contributed to its com-
pletion. The reviewers commissioned by Brooks/Cole and Cengage Learning made numerous
thoughtful and insightful suggestions that were incorporated into the final draft. Thanks to
graduate students Joe Tari, Rui Guo, and Lingyun Bai for helping review the initial FDSP tool-
box software. We would also like to thank a number of individuals at Brooks/Cole who helped
see this project to completion and mold the final product. Special thanks to Bill Stenquist who
worked closely with us throughout, and to Rose Kernan. The second edition from Cengage
Learning was made possible through the efforts and support of the dedicated group at Global
Engineering including Swati Meherishi, Hilda Gowans, Lauren Betsos, Tanya Altieri, and
Chris Shortt.
Robert J. Schilling
Sandra L. Harris
Potsdam, NY

Contents

Margin Contents xvii

PART I Signal and System Analysis 1

•1 •Signal
• • •Processing
• • • • •••••
3
••

1.1 Motivation 3
1.1.1 Digital and Analog Processing 4
1.1.2 Total Harmonic Distortion (THD) 6
1.1.3 A Notch Filter 7
1.1.4 Active Noise Control 7
1.1.5 Video Aliasing 10
1.2 Signals and Systems 11
1.2.1 Signal Classification 11
1.2.2 System Classification 16
1.3 Sampling of Continuous-time Signals 21
1.3.1 Sampling as Modulation 21
1.3.2 Aliasing 23
1.4 Reconstruction of Continuous-time Signals 26
1.4.1 Reconstruction Formula 26
1.4.2 Zero-order Hold 29
1.5 Prefilters and Postfilters 33
1.5.1 Anti-aliasing Filter 33
1.5.2 Anti-imaging Filter 37
∗
1.6 DAC and ADC Circuits 39
1.6.1 Digital-to-analog Converter (DAC) 39
1.6.2 Analog-to-digital Converter (ADC) 41
1.7 The FDSP Toolbox 46
1.7.1 FDSP Driver Module 46
1.7.2 Toolbox Functions 46
1.7.3 GUI Modules 49
1.8 GUI Software and Case Studies 52
1.9 Chapter Summary 60

∗
Sections marked with a ∗ contain more advanced or specialized material that can be skipped without loss of continuity.

x Contents

1.10 Problems 62
1.10.1 Analysis and Design 62
1.10.2 GUI Simulation 67
1.10.3 MATLAB Computation 68

•2 •Discrete-time
• • • • • • •Systems
•••••
in the Time Domain
••
70
2.1 Motivation 70
2.1.1 Home Mortgage 71
2.1.2 Range Measurement with Radar 72
2.2 Discrete-time Signals 74
2.2.1 Signal Classification 74
2.2.2 Common Signals 79
2.3 Discrete-time Systems 82
2.4 Difference Equations 86
2.4.1 Zero-input Response 87
2.4.2 Zero-state Response 90
2.5 Block Diagrams 94
2.6 The Impulse Response 96
2.6.1 FIR Systems 97
2.6.2 IIR Systems 98
2.7 Convolution 100
2.7.1 Linear Convolution 100
2.7.2 Circular Convolution 103
2.7.3 Zero Padding 105
2.7.4 Deconvolution 108
2.7.5 Polynomial Arithmetic 109
2.8 Correlation 110
2.8.1 Linear Cross-correlation 110
2.8.2 Circular Cross-correlation 114
2.9 Stability in the Time Domain 117
2.10 GUI Software and Case Studies 119
2.11 Chapter Summary 129
2.12 Problems 132
2.12.1 Analysis and Design 133
2.12.2 GUI Simulation 140
2.12.3 MATLAB Computation 142

•3 •Discrete-time
• • • • • • •Systems
•••••
in the Frequency Domain
••
145
3.1 Motivation 145
3.1.1 Satellite Attitude Control 146
3.1.2 Modeling the Vocal Tract 148
3.2 Z-transform Pairs 149
3.2.1 Region of Convergence 150
3.2.2 Common Z-transform Pairs 153
3.3 Z-transform Properties 157
3.3.1 General Properties 157
3.3.2 Causal Properties 162

Contents xi

3.4 Inverse Z-transform 164

3.4.1 Noncausal Signals 164
3.4.2 Synthetic Division 164
3.4.3 Partial Fractions 166
3.4.4 Residue Method 170
3.5 Transfer Functions 174
3.5.1 The Transfer Function 174
3.5.2 Zero-State Response 176
3.5.3 Poles, Zeros, and Modes 177
3.5.4 DC Gain 180
3.6 Signal Flow Graphs 181
3.7 Stability in the Frequency Domain 184
3.7.1 Input-output Representations 184
3.7.2 BIBO Stability 185
3.7.3 The Jury Test 188
3.8 Frequency Response 191
3.8.1 Frequency Response 191
3.8.2 Sinusoidal Inputs 193
3.8.3 Periodic Inputs 196
3.9 System Identification 198
3.9.1 Least-squares Fit 199
3.9.2 Persistently Exciting Inputs 202
3.10 GUI Software and Case Studies 203
3.10.1 g sysfreq: Discrete-time System Analysis
in the Frequency Domain 203
3.11 Chapter Summary 213
3.12 Problems 215
3.12.1 Analysis and Design 215
3.12.2 GUI Simulation 225
3.12.3 MATLAB Computation 226

•4 •Fourier
• • • •Transforms
• • • •••••
and Spectral Analysis
••
228
4.1 Motivation 228
4.1.1 Fourier Series 229
4.1.2 DC Wall Transformer 230
4.1.3 Frequency Response 232
4.2 Discrete-time Fourier Transform (DTFT) 233
4.2.1 DTFT 233
4.2.2 Properties of the DTFT 236
4.3 Discrete Fourier Transform (DFT) 241
4.3.1 DFT 241
4.3.2 Matrix Formulation 243
4.3.3 Fourier Series and Discrete Spectra 245
4.3.4 DFT Properties 248
4.4 Fast Fourier Transform (FFT) 256
4.4.1 Decimation in Time FFT 256
4.4.2 FFT Computational Effort 260
4.4.3 Alternative FFT Implementations 262

xii Contents

4.5 Fast Convolution and Correlation 263

4.5.1 Fast Convolution 263
∗
4.5.2 Fast Block Convolution 267
4.5.3 Fast Correlation 270
4.6 White Noise 274
4.6.1 Uniform White Noise 274
4.6.2 Gaussian White Noise 278
4.7 Auto-correlation 282
4.7.1 Auto-correlation of White Noise 282
4.7.2 Power Density Spectrum 284
4.7.3 Extracting Periodic Signals from Noise 286
4.8 Zero Padding and Spectral Resolution 291
4.8.1 Frequency Response Using the DFT 291
4.8.2 Zero Padding 295
4.8.3 Spectral Resolution 296
4.9 Spectrogram 299
4.9.1 Data Windows 299
4.9.2 Spectrogram 301
4.10 Power Density Spectrum Estimation 304
4.10.1 Bartlett’s Method 304
4.10.2 Welch’s Method 308
4.11 GUI Software and Case Studies 311
4.12 Chapter Summary 319
4.13 Problems 323
4.13.1 Analysis and Design 323
4.13.2 GUI Simulation 329
4.13.3 MATLAB Computation 331

PART II Digital Filter Design 335

•5 •Filter
• • •Design
• • • • •••••
Specifications
••
337
5.1 Motivation 337
5.1.1 Filter Design Specifications 338
5.1.2 Filter Realization Structures 339
5.2 Frequency-selective Filters 342
5.2.1 Linear Design Specifications 343
5.2.2 Logarithmic Design Specifications (dB) 348
5.3 Linear-phase and Zero-phase Filters 350
5.3.1 Linear Phase 350
5.3.2 Zero-phase Filters 356
5.4 Minimum-phase and Allpass Filters 358
5.4.1 Minimum-phase Filters 359
5.4.2 Allpass Filters 362
5.4.3 Inverse Systems and Equalization 366
5.5 Quadrature Filters 367
5.5.1 Differentiator 367
5.5.2 Hilbert Transformer 369
5.5.3 Digital Oscillator 372

Contents xiii

5.6 Notch Filters and Resonators 374

5.6.1 Notch Filters 374
5.6.2 Resonators 376
5.7 Narrowband Filters and Filter Banks 378
5.7.1 Narrowband Filters 378
5.7.2 Filter Banks 381
5.8 Adaptive Filters 383
5.9 GUI Software and Case Study 386
5.9.1 g filters: Evaluation of Digital Filter Characteristics 386
5.10 Chapter Summary 392
5.11 Problems 395
5.11.1 Analysis and Design 395
5.11.2 GUI Simulation 403
5.11.3 MATLAB Computation 404

•6 •FIR
• •Filter
• • • • • •••••
Design 406
••

6.1 Motivation 406

6.1.1 Numerical Differentiators 407
6.1.2 Signal-to-noise Ratio 409
6.2 Windowing Method 411
6.2.1 Truncated Impulse Response 412
6.2.2 Windowing 416
6.3 Frequency-sampling Method 424
6.3.1 Frequency Sampling 424
6.3.2 Transition-band Optimization 425
6.4 Least-squares Method 430
6.5 Equiripple Filters 434
6.5.1 Minimax Error Criterion 434
6.5.2 Parks-McClellan Algorithm 436
6.6 Differentiators and Hilbert Transformers 442
6.6.1 Differentiators 442
6.6.2 Hilbert Transformers 445
6.7 Quadrature Filters 448
6.7.1 Generation of a Quadrature Pair 448
6.7.2 Quadrature Filter 450
6.7.3 Equalizer Design 453
6.8 Filter Realization Structures 457
6.8.1 Direct Forms 457
6.8.2 Cascade Form 459
6.8.3 Lattice Form 461
∗
6.9 Finite Word Length Effects 464
6.9.1 Binary Number Representation 465
6.9.2 Input Quantization Error 466
6.9.3 Coefficient Quantization Error 470
6.9.4 Roundoff Error, Overflow, and Scaling 473
6.10 GUI Software and Case Study 477
6.11 Chapter Summary 484

xiv Contents

6.12 Problems 488

6.12.1 Analysis and Design 488
6.12.2 GUI Simulation 492
6.12.3 MATLAB Computation 494

•7 •IIR• •Filter
• • • • • •••••
Design 499
••

7.1 Motivation 499

7.1.1 Tunable Plucked-string Filter 500
7.1.2 Colored Noise 502
7.2 Filter Design by Pole-zero Placement 504
7.2.1 Resonator 504
7.2.2 Notch Filter 508
7.2.3 Comb Filters 510
7.3 Filter Design Parameters 514
7.4 Classical Analog Filters 517
7.4.1 Butterworth Filters 517
7.4.2 Chebyshev-I Filters 522
7.4.3 Chebyshev-II Filters 525
7.4.4 Elliptic Filters 526
7.5 Bilinear-transformation Method 529
7.6 Frequency Transformations 535
7.6.1 Analog Frequency Transformations 536
7.6.2 Digital Frequency Transformations 539
7.7 Filter Realization Structures 541
7.7.1 Direct Forms 541
7.7.2 Parallel Form 544
7.7.3 Cascade Form 547
∗
7.8 Finite Word Length Effects 550
7.8.1 Coefficient Quantization Error 550
7.8.2 Roundoff Error, Overflow, and Scaling 553
7.8.3 Limit Cycles 557
7.9 GUI Software and Case Study 560
7.10 Chapter Summary 567
7.11 Problems 571
7.11.1 Analysis and Design 571
7.11.2 GUI Simulation 575
7.11.3 MATLAB Computation 578

PART III Advanced Signal Processing 581

•8 •Multirate
• • • • •Signal
• • •••••
Processing
••
583
8.1 Motivation 583
8.1.1 Narrowband Filter Banks 584
8.1.2 Fractional Delay Systems 586
8.2 Integer Sampling Rate Converters 587
8.2.1 Sampling Rate Decimator 587
8.2.2 Sampling Rate Interpolator 588

Contents xv

8.3 Rational Sampling Rate Converters 591

8.3.1 Single-stage Converters 591
8.3.2 Multistage Converters 593
8.4 Multirate Filter Realization Structures 596
8.4.1 Polyphase Decimator 596
8.4.2 Polyphase Interpolator 598
8.5 Narrowband Filters and Filter Banks 600
8.5.1 Narrowband Filters 600
8.5.2 Filter Banks 601
8.6 A Two-channel QMF Bank 607
8.6.1 Rate Converters in the Frequency Domain 608
8.6.2 An Alias-free QMF Bank 610
8.7 Oversampling ADC 612
8.7.1 Anti-aliasing Filters 612
8.7.2 Sigma-delta ADC 615
8.8 Oversampling DAC 620
8.8.1 Anti-imaging Filters 620
8.8.2 Passband Equalization 621
8.9 GUI Software and Case Study 623
8.10 Chapter Summary 630
8.11 Problems 633
8.11.1 Analysis and Design 633
8.11.2 GUI Simulation 641
8.11.3 MATLAB Computation 642

•9 •Adaptive
• • • • Signal
• • • •••••
Processing
••
645
9.1 Motivation 645
9.1.1 System Identification 646
9.1.2 Channel Equalization 647
9.1.3 Signal Prediction 648
9.1.4 Noise Cancellation 648
9.2 Mean Square Error 649
9.2.1 Adaptive Transversal Filters 649
9.2.2 Cross-correlation Revisited 650
9.2.3 Mean Square Error 651
9.3 The Least Mean Square (LMS) Method 656
9.4 Performance Analysis of LMS Method 660
9.4.1 Step Size 660
9.4.2 Convergence Rate 663
9.4.3 Excess Mean Square Error 666
9.5 Modified LMS Methods 669
9.5.1 Normalized LMS Method 669
9.5.2 Correlation LMS Method 671
9.5.3 Leaky LMS Method 674
9.6 Adaptive FIR Filter Design 678
9.6.1 Pseudo-filters 678
9.6.2 Linear-phase Pseudo-filters 681

xvi Contents

9.7 The Recursive Least Squares (RLS) Method 684

9.7.1 Performance Criterion 684
9.7.2 Recursive Formulation 685
9.8 Active Noise Control 690
9.8.1 The Filtered-x LMS Method 691
9.8.2 Secondary Path Identification 693
9.8.3 Signal-synthesis Method 695
9.9 Nonlinear System Identification 700
9.9.1 Nonlinear Discrete-time Systems 700
9.9.2 Grid Points 701
9.9.3 Radial Basis Functions 703
9.9.4 Adaptive RBF Networks 707
9.10 GUI Software and Case Study 713
9.11 Chapter Summary 718
9.12 Problems 721
9.12.1 Analysis and Design 721
9.12.2 GUI Simulation 726
9.12.3 MATLAB Computation 727

•References
• • • • •and
• • • •••••
Further Reading
••
734

•Appendix
• • • •1• •Transform
• • •••••
Tables
••
738
1.1 Fourier Series 738
1.2 Fourier Transform 739
1.3 Laplace Transform 741
1.4 Z-transform 743
1.5 Discrete-time Fourier Transform 744
1.6 Discrete Fourier Transform (DFT) 745

•Appendix
• • • •2• •Mathematical
• • •••••
Identities
••
747
2.1 Complex Numbers 747
2.2 Euler’s Identity 747
2.3 Trigonometric Identities 748
2.4 Inequalities 748
2.5 Uniform White Noise 749

•Appendix
• • • •3• •FDSP
• • •••••
Toolbox Functions
••
750
3.1 Installation 750
3.2 Driver Module: f dsp 751
3.3 Chapter GUI Modules 751
3.4 FDSP Toolbox Functions 752

•Index
• • •755• • • • • • • • • • ••

Margin Contents

TABLE I: Number Term Symbol Page

Definitions
1.1 Causal signal xa (t) 15
1.2 Linear system S 17
1.3 Time-invariant system S 17
1.4 Stable system S 18
1.5 Frequency response Ha ( f) 19
1.6 Impulse response ha (t) 20
1.7 Bandlimited signal xa (t) 24
1.8 Transfer function Ha (s) 29

2.1 Impulse response h(k) 97

2.2 FIR and IIR systems S 97
2.3 Linear convolution h(k) x(k) 101
2.4 Circular convolution h(k) ◦ x(k) 104
2.5 Linear cross-correlation r yx (k) 111
2.6 Circular cross-correlation c yz(k) 114
2.7 BIBO stable h1 < ∞ 117

3.1 Z-transform X(z) 149

3.2 Transfer function H (z) 174
3.3 Frequency response H (f) 191

4.1 Discrete-time Fourier transform (DTFT) X(f) 233

4.2 Discrete Fourier transform (DFT) X(i) 242
4.3 Expected value E [f(x)] 275
4.4 Circular auto-correlation cxx (k) 282
4.5 Spectrogram G(m, i) 301

5.1 Group delay D(f) 351

5.2 Linear-phase filter H (z) 351
5.3 Minimum-phase filter H (z) 359
5.4 Allpass Filter H (z) 362

6.1 Signal-to-noise ratio SNR(y) 410

6.2 Quantization operator Q N (x) 466

9.1 Random cross-correlation r yx (i) 651

xvii

xviii Margin Contents

TABLE II: Number Description Page

Propositions
1.1 Signal Sampling 25
1.2 Signal Reconstruction 28

2.1 BIBO Stability: Time Domain 118

3.1 BIBO Stability: Frequency Domain 186

3.2 Frequency Response 194

4.1 Parseval’s Identity: DTFT 238

4.2 Parseval’s Identity: DFT 254

5.1 Paley-Wiener Theorem 344

5.2 Linear-phase Filter 353
5.3 Minimum-phase Allpass Decomposition 363

6.1 Alternation Theorem 436

6.2 Flow Graph Reversal Theorem 458

9.1 LMS Convergence 661

TABLE III: Number Description Page

Algorithms
1.1 Successive Approximation 43

3.1 Residue Method 172

4.1 Bit Reversal 258

4.2 FFT 260
4.3 Problem Domain 261
4.4 IFFT 262
4.5 Fast Block Convolution 268

5.1 Zero-phase Filter 357

5.2 Minimum-phase Allpass Decomposition 364

6.1 Windowed FIR Filter 421

6.2 Equiripple FIR Filter 438
6.3 Lattice-form Realization 462

7.1 Bilinear Transformation Method 532

9.1 RLS Method 687

9.2 RBF Network Evaluation 708

1
2
3
4
5
6
7
8
9
10
11 This page was intentionally left blank
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
S 50
R 51

1st Pass Pages

Schilling-1120949 book November 12, 2010 9:33

PART I

Signal and System Analysis

1
Signal Processing

2 3
Discrete Systems, Discrete Systems,
Time Domain Frequency Domain

4
Fourier Transforms,
Signal Spectra

1
Schilling-1120949 book November 12, 2010 9:33

CHAPTER

1 Signal Processing

•••••
• • • • • •• • • •••••
Chapter Topics
1.1 Motivation
1.2 Signals and Systems
1.3 Sampling of Continuous-time Signals
1.4 Reconstruction of Continuous-time Signals
1.5 Prefilters and Postfilters
1.6 DAC and ADC Circuits
1.7 The FDSP Toolbox
1.8 GUI Software and Case Study
1.9 Chapter Summary
1.10 Problems

• • • • • • • • • ••••• ••

1.1 Motivation
A signal is a physical variable whose value varies with time or space. When the value of
Continuous-time the signal is available over a continuum of time it is referred to as a continuous-time or
signal analog signal. Everyday examples of analog signals include temperature, pressure, liquid level,
chemical concentration, voltage and current, position, velocity, acceleration, force, and torque.
Discrete-time signal If the value of the signal is available only at discrete instants of time, it is called a discrete-
time signal. Although some signals, for example economic data, are inherently discrete-time
signals, a more common way to produce a discrete-time signal, x(k), is to take samples of an
underlying analog signal, xa (t).

x(k) = xa (kT ), |k| = 0, 1, 2, · · ·

Sampling interval Here T denotes the sampling interval or time between samples, and = means equals by
definition. When finite precision is used to represent the value of x(k), the sequence of quantized
Digital signal values is then called a digital signal. A system or algorithm which processes one digital signal
x(k) as its input and produces a second digital signal y(k) as its output is a digital signal
processor. Digital signal processing (DSP) techniques have widespread applications, and they
play an increasingly important role in the modern world. Application areas include speech

4 Chapter 1 Signal Processing

recognition, detection of targets with radar and sonar, processing of music and video, seismic
exploration for oil and gas deposits, medical signal processing including EEG, EKG, and
ultrasound, communication channel equalization, and satellite image processing. The focus of
this book is the development, implementation, and application of modern DSP techniques.
We begin this introductory chapter with a comparison of digital and analog signal process-
ing. Next, some practical problems are posed that can be solved using DSP techniques. This
is followed by characterization and classification of signals. The fundamental notion of the
spectrum of a signal is then presented including the concepts of bandlimited and white-noise
signals. This leads naturally to the sampling process which takes a continuous-time signal and
produces a corresponding discrete-time signal. Simple conditions are presented that ensure
that an analog signal can be reconstructed from its samples. When these conditions are not
satisfied, the phenomenon of aliasing occurs. The use of guard filters to reduce the effects of
aliasing is discussed. Next DSP hardware in the form of analog-to-digital converters (ADCs)
and digital-to-analog converters (DACs) is examined. The hardware discussion includes ways
to model the quantization error associated with finite precision converters. A custom MATLAB
toolbox, called FDSP, is then introduced that facilitates the development of simple DSP pro-
GUI modules grams. The FDSP toolbox also includes a number of graphical user interface (GUI) modules
that can be used to browse examples and explore digital signal processing techniques without
any need for programming. The GUI module g sample allows the user to investigate the sig-
nal sampling process, while the companion module g reconstruct allows the user to explore
the signal reconstruction process. The chapter concludes with a case study example, and a
summary of continuous-time and discrete-time signal processing.

1.1.1 Digital and Analog Processing

For many years, almost all signal processing was done with analog circuits as shown in
Figure 1.1. Here, operational amplifiers, resistors, and capacitors are used to realize frequency
selective filters.
With the advent of specialized microprocessors with built-in data conversion circuits
(Papamichalis, 1990), it is now commonplace to perform signal processing digitally as shown
in Figure 1.2. Digital processing of analog signals is more complex because it requires, at a
minimum, the three components shown in Figure 1.2. The analog-to-digital converter or ADC
at the front end converts the analog input xa (t) into an equivalent digital signal x(k). The

FIGURE 1.1: Analog

Signal Processing Analog
xa (t) e - processing e ya (t)
circuit

x(k) Digital y(k)

xa (t) e - ADC - processing - DAC e ya (t)
program

FIGURE 1.2: Digital Signal Processing

1.1 Motivation 5

TABLE 1.1: Feature Analog Processing Digital Processing

Comparison of
Analog and Digital Speed Fast Moderate
Cost Low to moderate Moderate
Signal Processing
Flexibility Low High
Performance Moderate High
Self-calibration No Yes
Data-logging capability No Yes
Adaptive capability Limited Yes

processing of x(k) is then achieved with an algorithm that is implemented in software. For
a filtering operation, the DSP algorithm consists of a difference equation, but other types of
processing are also possible and are often used. The digital output signal y(k) is then converted
back to an equivalent analog signal ya (t) by the digital-to-analog converter or DAC.
Although the DSP approach requires more steps than analog signal processing, there are
many important benefits to working with signals in digital form. A comparison of the relative
advantages and disadvantages of the two approaches is summarized in Table 1.1. Although the
DSP approach requires more steps than analog signal processing, there are many important
benefits to working with signals in digital form. A comparison of the relative advantages and
disadvantages of the two approaches is summarized in Table 1.1.
The primary advantages of analog signal processing are speed and cost. Digital signal
processing is not as fast due to the limits on the sampling rates of the converter circuits. In
addition, if substantial computations are to be performed between samples, then the clock rate
Real time of the processor also can be a limiting factor. Speed can be an issue in real-time applications
where the kth output sample y(k) must be computed and sent to the DAC as soon as possible
after the kth input sample x(k) is available from the ADC. However, there are also applications
where the entire input signal is available ahead of time for processing off-line. For this batch
mode type of processing, speed is less critical.
DSP hardware is often somewhat more expensive than analog hardware because analog
hardware can consist of as little as a few discrete components on a stand-alone printed circuit
board. The cost of DSP hardware varies depending on the performance characteristics required.
In some cases, a PC may already be available to perform other functions for a given application,
and in these instances the marginal expense of adding DSP hardware is not large.
In spite of these limitations, there are great benefits to using DSP techniques. Indeed, DSP
is superior to analog processing with respect to virtually all of the remaining features listed
in Table 1.1. One of the most important advantages is the inherent flexibility available with a
software implementation. Whereas an analog circuit might be tuned with a potentiometer to
vary its performance over a limited range, the DSP algorithm can be completely replaced, on
the fly, when circumstances warrant.
DSP also offers considerably higher performance than analog signal processing. For ex-
ample, digital filters with arbitrary magnitude responses and linear phase responses can be
designed easily whereas this is not feasible with analog filters.
A common problem that plagues analog systems is the fact that the component values tend
to drift with age and with changes in environmental conditions such as temperature. This leads
to a need for periodic calibration or tuning. With DSP there is no drift problem and therefore
no need to manually calibrate.
Since data are already available in digital form in a DSP system, with little or no additional
expense, one can log the data associated with the operation of the system so that its performance
can be monitored, either locally of remotely over a network connection. If an unusual operating
condition is detected, its exact time and nature can be determined and a higher-level control

6 Chapter 1 Signal Processing

FIGURE 1.3: An
Audio Amplifier
xa (t) e - K e ya (t)

system can be alerted. Although strip chart recorders can be added to an analog system, this
substantially increases the expense thereby negating one of its potential advantages.
The flexibility inherent in software can be exploited by having the parameters of the DSP
algorithm vary with time and adapt as the characteristics of the input signal or the processing
task change. Applications, like system identification and active noise control, exploit adaptive
signal processing, a topic that is addressed in Chapter 9.

1.1.2 Total Harmonic Distortion (THD)

With the widespread use of digital computers, DSP applications are now commonplace. As a
simple initial example, consider the problem of designing an audio amplifier to boost signal
strength without distorting the shape of the input signal. For the amplifier shown in Figure 1.3,
suppose the input signal xa (t) is a pure sinusoidal tone of amplitude a and frequency F0 Hz.

xa (t) = a cos(2π F0 t) (1.1.1)

An ideal amplifier will produce a desired output signal yd (t) that is a scaled and delayed version
of the input signal. For example, if the scale factor or amplifier gain is K and the delay is τ ,
then the desired output is

yd (t) = K xa (t − τ )
= K a cos[2π F0 (t − τ )] (1.1.2)

In a practical amplifier, the relationship between the input and the output is only approximately
linear, so some additional terms are present in the actual output ya .

ya (t) = F[xa (t)]

d0
M−1
≈ + di cos(2πi F0 t + θi ) (1.1.3)
2 i=1

The presence of the additional harmonics indicates that there is distortion in the amplified
signal due to nonlinearities within the amplifier. For example, if the amplifier is driven with
an input whose amplitude a is too large, then the amplifier will saturate with the result that the
output is a clipped sine wave that sounds distorted when played through a speaker. To quantify
the amount of distortion, the average power contained in the ith harmonic is di2 /2 for i ≥ 1
and di2 /4 for i = 0. Thus the average power of the signal ya (t) is

1 2
M−1
d02
Py = + d (1.1.4)
4 2 i=1 i

Total harmonic The total harmonic distortion or THD of the output signal ya (t) is defined as the power
distortion in the spurious harmonic components, expressed as a percentage of the total power. Thus the

1.1 Motivation 7

following can be used to measure the quality of the amplifier output.

100(Py − d12 /2)
THD = % (1.1.5)
Py
For an ideal amplifier di = 0 for i = 1, and
d1 = K a (1.1.6a)
θ1 = −2π F0 τ (1.1.6b)
Consequently, for a high-quality amplifier, the THD is small, and when no distortion is present
THD = 0. Suppose the amplifier output is sampled to produce the following digital signal of
length N = 2M.
y(k) = ya (kT ), 0 ≤ k < N (1.1.7)
If the sampling interval is set to T = 1/(N F0 ), then this corresponds to one period of ya (t). By
processing the digital signal x(k) with the discrete Fourier transform or DFT, it is possible to
determine di and θi for 0 ≤ i < M. In this way the total harmonic distortion can be measured.
The DFT is a key analytic tool that is introduced in Chapter 4.

1.1.3 A Notch Filter

As a second example of a DSP application, suppose one is performing sensitive acoustic
measurements in a laboratory setting using a microphone. Here, any ambient background
sounds in the range of frequencies of interest have the potential to corrupt the measurements
with unwanted noise. Preliminary measurements reveal that the overhead fluorescent lights are
emitting a 120 Hz hum which corresponds to the second harmonic of the 60 Hz commercial
AC power. The problem then is to remove the 120 Hz frequency component while affecting the
other nearby frequency components as little as possible. Consequently, you want to process the
Notch filter acoustic data samples with a notch filter designed to remove the effects of the fluorescent lights.
After some calculations, you arrive at the following digital filter to process the measurements
x(k) to produce a filtered signal y(k).
y(k) = 1.6466y(k − 1) − .9805y(k − 2) + .9905x(k)
−1.6471x(k − 1) + .9905x(k − 2) (1.1.8)
The filter in (1.1.8) is a notch filter with a bandwidth of 4 Hz, a notch frequency of Fn = 120
Hz, and a sampling frequency of f s = 1280 Hz. A plot of the frequency response of this
filter is shown in Figure 1.4 where a sharp notch at 120 Hz is apparent. Notice that except
for frequencies near Fn , all other frequency components of x(k) are passed through the filter
without attenuation. The design of notch filters is discussed in Chapter 7.

1.1.4 Active Noise Control

An application area of DSP that makes use of adaptive signal processing is active control
of acoustic noise (Kuo and Morgan, 1996). Examples include industrial noise from rotating
machines, propeller and jet engine noise, road noise in an automobile, and noise caused by
air flow in heating, ventilation, and air conditioning systems. As an illustration of the latter,
consider the active noise control system shown in Figure 1.5 which consists of an air duct
with two microphones and a speaker. The basic principle of active noise control is to inject
a secondary sound into the environment so as to cancel the primary sound using destructive
interference.

8 Chapter 1 Signal Processing

FIGURE 1.4: Magnitude Response

Magnitude 1.4
Response of a
Notch Filter with
1.2
F n = 120 Hz

0.8
A(f)

0.6

0.4

0.2

0
0 100 200 300 400 500 600 700
f (Hz)

FIGURE 1.5: Active

Control of Acoustic
Noise in an Air Duct Blower
Reference Speaker Error
microphone microphone
e @ e
@

y(k)

x(k) e(k)
- Controller

The purpose of the reference microphone in Figure 1.5 is to detect the primary noise x(k)
generated by the noise source or blower. The primary noise signal is then passed through a
digital filter of the following form.

m
y(k) = wi (k)x(k − i) (1.1.9)
i=0

The output of the filter y(k) drives a speaker that creates the secondary sound sometimes called
Antisound antisound. The error microphone, located downstream of the speaker, detects the sum of the pri-
mary and secondary sounds and produces an error signal e(k). The objective of the adaptive
algorithm is the take x(k) and e(k) as inputs and adjust the filter weights w(k) so as to drive
e2 (k) to zero. If zero error can be achieved, then silence is observed at the error microphone. In
practical systems, the error or residual sound is significantly reduced by active noise control.

1.1 Motivation 9

To illustrate the operation of this adaptive DSP system, suppose the blower noise is modeled
as a periodic signal with fundamental frequency F0 and r harmonics plus some random white
noise v(k).

r
x(k) = ai cos(2πik F0 T + θi ) + v(k), 0 ≤ k < p (1.1.10)
i=1

For example, suppose F0 = 100 Hz and there are r = 4 harmonics with amplitudes ai = 1/i
and random phase angles. Suppose the random white noise term is distributed uniformly over
the interval [−.5, .5]. Let p = 2048 samples, suppose the sampling interval is T = 1/1600
sec, and the filter order is m = 40. The adaptive algorithm used to adjust the filter weights is
called the FXLMS method, and it is discussed in detail in Chapter 9. The results of applying
this algorithm are shown in Figure 1.6.
Initially the filter weights are set to w(0) = 0 which corresponds to no noise control at all.
The adaptive algorithm is not activated until sample k = 512, so the first quarter of the plot
in Figure 1.6 represents the ambient or primary noise detected at the error microphone. When
adaptation is activated, the error begins to decrease rapidly and after a short transient period
it reaches a steady-state level that is almost two orders of magnitude quieter than the primary
noise itself. We can quantify the noise reduction by using the following measure of overall
noise cancellation.
p/4−1 p−1

E = 10 log10 e (i) − 10 log10
2
e (i) dB
2
(1.1.11)
i =0 i = 3 p/4

The overall noise cancellation E is the log of the ratio of the average power of the noise during
the first quarter of the samples divided by the average power of the noise during the last quarter
of the samples, expressed in units of decibels. Using this measure, the noise cancellation
observed in Figure 1.6 is E = 37.8 dB.

FIGURE 1.6: Error Squared Error

Signal with Active 100
Noise Control
Activated at 90
k = 512
80

60
e2(k)

Noise reduction = 37.8 dB

0
0 500 1000 1500 2000
k

10 Chapter 1 Signal Processing

1.1.5 Video Aliasing

Later in Chapter 1 we focus on the problem of sampling a continuous-time signal xa (t) to
produce the following discrete-time signal where T > 0 is the sampling interval and f s = 1/T
is the sampling frequency.
x(k) = xa (kT ), |k| = 0, 1, 2, . . . (1.1.12)
An important theoretical and practical question that arises in connection with the sampling
process is this: under what conditions do the samples x(k) contain all the information needed
to reconstruct the signal xa (t)? The Shannon sampling theorem (Proposition 1.1), says that if
the signal xa (t) is bandlimited and the sampling rate f s is greater than twice the bandwidth
or highest frequency present, then it is possible to interpolate between the x(k) to precisely
reconstruct xa (t). However, if the sampling frequency is too low, then the samples become
Aliasing corrupted, a process known as aliasing. An easy way to interpret aliasing is to examine a
video signal in the form of an M × N image Ia (t) that varies with time. Here Ia (t) consists of
Pixels an M × N array of picture elements or pixels where the number of rows M and columns N
depends on the video format used. If Ia (t) is sampled with a sampling interval of T then the
resulting M N -dimensional discrete-time signal is
I (k) = Ia (kT ), |k| = 0, 1, 2, . . . (1.1.13)
Here, f s = 1/T is the sampling rate in frames/second. Depending on the content of the image,
the sampling rate f s may or may not be sufficiently high to avoid aliasing.
As a simple illustration, suppose the image consists of a rotating disk with a dark line on
it to indicate orientation as shown in Figure 1.7. A casual look at the sequence of frames in
Figure 1.7 suggests that the disk appears to be rotating counterclockwise at a rate of 45 degrees
per frame. However, this is not the only interpretation possible. For example, an alternative
FIGURE 1.7: Four k = 0 k = 1
Video Frames of a 5 5
Rotating Disk

0 0

−5 −5
−5 0 5 −5 0 5

k = 2 k = 3
5 5

0 0

−5 −5
−5 0 5 −5 0 5

1.2 Signals and Systems 11

explanation is that the disk is actually rotating clockwise at a rate of 315 degrees/frame.
Both interpretations are plausible. Is the motion captured by the snapshots a fast clockwise
rotation or a slow counter clockwise rotation? If the disk is in fact rotating clockwise at F0
revolutions/second, but the sampling rate is f s ≤ 2F0 , then aliasing occurs in which case the
disk can appear to turn backwards at a slow rate. Interestingly, this manifestation of aliasing
was quite common in older western films that featured wagon trains heading west. The spokes
on the wagon wheels sometimes appeared to move backwards because of the slow frame rate
used to shoot the film and display it on older TVs.

• • • • • • • • • ••••• ••

1.2 Signals and Systems

1.2.1 Signal Classification
Recall that a signal is a physical variable whose value varies with respect to time or space.
To simplify the notation and terminology, we will assume that, unless noted otherwise, the
independent variable denotes time. If the value of the signal, the dependent variable, is available
Continuous-time over a continuum of times, t ∈ R, then the signal is referred to as a continuous-time signal.
signal An example of a continuous-time signal, xa (t), is shown in Figure 1.8.
In many cases of practical interest, the value of the signal is only available at discrete
Discrete-time signal instants of time in which case it is referred to as a discrete-time signal. That is, signals can be
classified into continuous-time or discrete-time depending on whether the independent variable
is continuous or discrete, respectively. Common everyday examples of discrete-time signals
include economic statistics such as the daily balance in one’s savings account, or the monthly
inflation rate. In DSP applications, a more common way to produce a discrete-time signal,
x(k), is to sample an underlying continuous-time signal, xa (t), as follows.
x(k) = xa (kT ), |k| = 0, 1, 2, · · · (1.2.1)

FIGURE 1.8: A xa(t) = 10t exp(−t)

Continuous-time 4
Signal xa (t)
3.5

2.5
x (t)

2
a

1.5

0.5

0
0 0.5 1 1.5 2 2.5 3 3.5 4
t (sec)

12 Chapter 1 Signal Processing

Sampling interval Here, T > 0 is the time between samples or sampling interval in seconds. The sample spacing
Sampling frequency also can be specified using the reciprocal of the sampling interval which is called the sampling
frequency, f s .
1
Hz fs = (1.2.2)
T
Here, the unit of Hz is understood to mean samples/second. Notice that the integer k in (1.2.1)
Discrete-time denotes discrete time or, more specifically, the sample number. The sampling interval T is
left implicit on the left-hand side of (1.2.1) because this simplifies subsequent notation. In
those instances where the value of T is important, it will be stated explicitly. An example of
a discrete-time signal generated by sampling the continuous-time signal in Figure 1.8 using
T = .25 seconds is shown in Figure 1.9.
Just as the independent variable can be continuous or discrete, so can the dependent variable
or amplitude of the signal be continuous or discrete. If the number of bits of precision used to
Quantized signal represent the value of x(k) is finite, then we say that x(k) is a quantized or discrete-amplitude
signal. For example, if N bits are used to represent the value of x(k), then there are 2 N distinct
values that x(k) can assume. Suppose the value of x(k) ranges over the interval [xm , x M ]. Then
Quantization level the quantization level, or spacing between adjacent discrete values of x(k), is
x M − xm
q= (1.2.3)
2N
The quantization process can be thought of as passing a signal through a piecewise-constant
staircase type function. For example, if the quantization is based on rounding to the nearest N
Quantization bits, then the process can be represented with the following quantization operator.
operator
x
Q N (x) = q · round (1.2.4)
q
A graph of Q N (x) for x ranging over the interval [−1, 1] using N = 5 bits is shown in
Digital signal Figure 1.10. A quantized discrete-time signal is called a digital signal. That is, a digital signal,

FIGURE 1.9: A x(k) = 10kT exp(−kT)

Discrete-time Signal 4
x(k) with T = .25
3.5

2.5
x(k)

1.5 T = .25

0.5

0
0 .5 1 1.5 2 2.5 3 3.5 4
kT (sec)

1.2 Signals and Systems 13

FIGURE 1.10: Quantizer Input−output Characteristic

Quantization over 1
[−1, 1] Using N = 5
Bits 0.8

0.6
q = .0625
0.4

0.2
Q(x)
0

−0.2

−0.4

−0.6

−0.8

−1
−1 −0.5 0 0.5 1
x

xq (k), is discrete in both time and amplitude with

xq (k) = Q N [xa (kT )] (1.2.5)
Analog signal By contrast, a signal that is continuous in both time and amplitude is called an analog signal.
An example of a digital signal obtained by quantizing the amplitude of the discrete-time signal
in Figure 1.9 is shown in Figure 1.11. In this case, the 5-bit quantizer in Figure 1.10 is used to

FIGURE 1.11: A xq(k) = QN[xa(kT)]

Digital Signal xq (k) 4

3.5

2.5
xq(k)

2
T = .25
1.5
q = .1290
N = 5 bits
1

0.5

0
0 .5 1 1.5 2 2.5 3 3.5 4
kT (sec)

14 Chapter 1 Signal Processing

produce xq (k). Careful inspection of Figure 1.11 reveals that at some of the samples there are
noticeable differences between xq (k) and xa (kT ). If rounding is used, then the magnitude of
the error is, at most, q/2.
Most of the analysis in this book will be based on discrete-time signals rather than digital
signals. That is, infinite precision is used to represent the value of the dependent variable.
Finite precision, or finite word length effects, are examined in Chapters 6 and 7 in the context
of digital filter design. When digital filter are implemented in MATLAB using the default
double-precision arithmetic, this corresponds to 64 bits of precision (16 decimal digits). In
most instances this is sufficiently high precision to yield insignificant finite word length effects.
Quantization A digital signal xq (k) can be modeled as a discrete-time signal x(k) plus random quanti-
noise zation noise, v(k), as follows.

xq (k) = x(k) + v(k) (1.2.6)

An effective way to measure the size or strength of the quantization noise is to use average
Expected value power defined as the mean, or expected value, of v 2 (k). Typically, v(k) is modeled as a
random variable uniformly distributed over the interval [−q/2, q/2] with probability density
p(x) = 1/q. In this case, the expected value of v 2 (k) is
q/2
E[v 2 ] = p(x)x 2 dx
−q/2
q/2
1
= x 2 dx (1.2.7)
q −q/2

Thus, the average power of the quantization noise is proportional to the square of the quanti-
zation level with

q2
E[v 2 ] = (1.2.8)
12

Example 1.1 Quantization Noise

Suppose the value of a discrete-time signal x(k) is constrained to lie in the interval [−10, 10].
Let xq (k) denote a digital version of x(k) using quantization level q, and consider the following
problem. Suppose the average power of the quantization noise, v(k), is to be less than .001.
What is the minimum number of bits that are needed to represent the value of xq (k)? The
constraint on the average power of the quantization noise is

E[v 2 ] < .001

Thus, from (1.2.3) and (1.2.8), we have

(x M − xm )2
< .001
12(2 N )2
Recall that the signal range is xm = −10 and x M = 10. Multiplying both sides by 12, taking
the square root of both sides, and then solving for 2 N yields
20
2N > √
.012

1.2 Signals and Systems 15

Finally, taking the natural log of both sides and solving for N we have
ln(182.5742)
= 7.5123
N >
ln(2)
Since N must be an integer, the minimum number of bits needed to ensure that the average
power of the quantization noise is less than .001 is N = 8 bits.

Signals can be further classified depending on whether or not they are nonzero for negative
values of the independent variable.
DEFINITION
A signal xa (t) defined for t ∈ R is causal if and only if it is zero for negative t. Otherwise,
1.1: Causal Signal
the signal is noncausal.
xa (t) = 0 for t < 0

Most of the signals that we work with will be causal signals. A simple, but important,
Unit step example of a causal signal is the unit step which is denoted μa (t) and defined

0, t < 0
μa (t) = (1.2.9)
1, t ≥ 0
Note that any signal can be made into a causal signal by multiplying by the unit step. For
example, xa (t) = exp(−t/τ )μa (t) is a causal decaying exponential with time constant τ .
Unit impulse Another important example of a causal signal is the unit impulse which is denoted δa (t).
Strictly speaking, the unit impulse is not a function because it is not defined at t = 0. However,
the unit impulse can be defined implicitly by the equation
t
δa (τ )dτ = μa (t) (1.2.10)
−∞

That is, the unit impulse δa (t) is a signal that, when integrated, produces the unit step μa (t).
Consequently, we can loosely think of the unit impulse as the derivative of the unit step function,
keeping in mind that the derivative of the unit step is not defined at t = 0. The two essential
characteristics of the unit impulse that follow from (1.2.10) are
δa (t) = 0, t = 0 (1.2.11a)
∞
δa (t)dt = 1 (1.2.11b)
−∞
A more informal way to view the unit impulse is to consider a narrow pulse of width and
height 1/ starting at t = 0. The unit impulse can be thought of as the limit of this sequence of
pulses as the pulse width goes to zero. By convention, we graph the unit impulse as a vertical
arrow with the height of the arrow equal to the strength, or area, of the impulse as shown in
Figure 1.12.
The unit impulse has an important property that is a direct consequence of (1.2.11). If xa (t)
is a continuous function, then
∞ ∞
xa (τ )δa (τ − t0 )dτ = xa (t0 )δa (τ − t0 )dτ
−∞ −∞
∞
= xa (t0 ) δa (τ − t0 )dτ
−∞∞
= xa (t0 ) δa (α)dα (1.2.12)
−∞

16 Chapter 1 Signal Processing

FIGURE 1.12: Unit Elementary Signals

Impulse, δa (t), and 1.5
Unit Step, μa (t)

ua(t )
1

xa(t)

0.5 d a(t )

−0.5
−2 −1.5 −1 −0.5 0 0.5 1 1.5 2
t (sec)

FIGURE 1.13: A
System S with Input
x and Output y x e - S e y

Sifting property Since the area under the unit impulse is one, we then have the following sifting property of the
unit impulse

∞
xa (t)δa (t − t0 )dt = xa (t0 ) (1.2.13)
−∞

From (1.2.13) we see that when a continuous function of time is multiplied by an impulse
and then integrated, the effect is to sift out or sample the value of the function at the time the
impulse occurs.

1.2.2 System Classification

Just as signals can be classified, so can the systems that process those signals. Consider a
system S with input x and output y as shown in Figure 1.13. In some instances, for example
biomedical systems, the input is referred to as the stimulus, and the output is referred to as the
response. We can think of the system in Figure 1.13 as an operator S that acts on the input
signal x to produce the output signal y.
y = Sx (1.2.14)
Continuous, discrete If the input and output are continuous-time signals, then the system S is called a continuous-
systems time system. A discrete-time system is a system S that processes a discrete-time input x(k)

1.2 Signals and Systems 17

to produce a discrete-time output y(k). There are also examples of systems that contain both
continuous-time signals and discrete-time signals. These systems are referred to as sampled-
data systems.
Almost all of the examples of systems in this book belong to an important class of systems
called linear systems.

DEFINITION
Let x1 and x2 be arbitrary inputs and let a and b be arbitrary scalars. A system S is linear
1.2: Linear System
if and only if the following holds, otherwise it is a nonlinear system.
S(ax1 + bx2 ) = aSx1 + bSx2

Thus a linear system has two distinct characteristics. When a = b = 1, we see that the response
to a sum of inputs is just the sum of the responses to the individual inputs. Similarly, when
b = 0, we see that the response to a scaled input is just the scaled response to the original input.
Examples of linear discrete-time systems include the notch filter in (1.1.8) and the adaptive
filter in (1.1.9). On the other hand, if the analog audio amplifier in Figure 1.3 is over driven and
its output saturates to produce harmonics as in (1.1.3), then this is an example of a nonlinear
continuous-time system. Another important class of systems is time-invariant systems.

DEFINITION
A system S with input xa (t) and output ya (t) is time-invariant if and only if whenever the
1.3: Time-invariant
input is translated in time by τ , the output is also translated in time by τ . Otherwise the
System
system is a time-varying system.
Sxa (t − τ ) = ya (t − τ )

For a time-invariant system, delaying or advancing the input delays or advances the output
by the same amount, but it does not otherwise affect the shape of the output. Therefore the results
of an input-output experiment do not depend on when the experiment is performed. Time-
invariant systems described by differential or difference equations have constant coefficients.
More generally, physical time-invariant systems have constant parameters. The notch filter in
(1.1.8) is an example of a discrete-time system that is both linear and time-invariant. On the
other hand, the adaptive digital filter in (1.1.9) is a time-varying system because the weights
w(k) are coefficients that change with time as the system adapts. The following example shows
that the concepts of linearity and time-invariance can sometimes depend on how the system is
characterized.

Example 1.2 System Classification

Consider the operational amplifier circuit shown in Figure 1.14. Here input resistor R1 is fixed,
but feedback resistor R2 represents a sensor or transducer whose resistance changes with
respect to a sensed environmental variable such as temperature or pressure. For this inverting
amplifier configuration, the output voltage ya (t) is
R2 (t)
ya (t) = − x1 (t)
R1
This is an example of a linear continuous-time system that is time-varying because parameter
R2 (t) varies as the temperature or pressure changes. However, another way to model this

18 Chapter 1 Signal Processing

FIGURE 1.14: An R2(t)

Inverting Amplifier
with a Feedback
Transducer R1
• −
+
•
+
x1 +
ya
− −
•
system is to consider the variable resistance of the sensor as a second input x2 (t) = R2 (t).
Viewing the system in this way, the system output is
x1 (t)x2 (t)
ya (t) = −
R1
This formulation of the model is a nonlinear time-invariant system, but with two inputs. Thus,
by introducing a second input we have converted a single-input time-varying linear system to
a two-input time-invariant nonlinear system.

Another important classification of systems focuses on the question of what happens to

Bounded signal the signals as time increases. We say that a signal xa (t) is bounded if and only if there exists a
Bx > 0 called a bound such that
|xa (t)| ≤ Bx for t ∈ R (1.2.15)
DEFINITION
A system S is with input xa (t) and output ya (t) is stable, in a bounded input bounded
1.4: Stable System
output (BIBO) sense, if and only if every bounded input produces a bounded output.
Otherwise it is an unstable system.

Thus an unstable system is a system for which the magnitude of the output grows arbitrarily
large with time for a least one bounded input.

Example 1.3 Stability

As a simple example of a system that can be stable or unstable depending on its parameter
values, consider the following first-order linear continuous-time system where a = 0.
dya (t)
+ aya (t) = xa (t)
dt
Suppose the input is the unit step xa (t) = μa (t) which is bounded with a bound of Bx = 1.
Direct substitution can be used to verify that for t ≥ 0, the solution is
1
1 − exp(−at)
ya (t) = ya (0) exp(−at) +
a
If a > 0, then the exponential terms grow without bound which means that the bounded input
u a (t) produces an unbounded output ya (t). Thus this system is unstable, in a BIBO sense,
when a > 0.

1.2 Signals and Systems 19

Just as light can be decomposed into a spectrum of colors, signals also contain energy that
is distributed over a range of frequencies. To decompose a continuous-time signal xa (t) into
Fourier transform its spectral components, we use the Fourier transform.
∞

X a ( f ) = F{xa (t)} = xa (t) exp(− j2π f t)dt (1.2.16)
−∞

It is assumed that the reader is familiar with the basics of continuous-time transforms, specifi-
cally the Laplace transform and the Fourier transform. Tables of transform pairs and transform
properties for all of the tranforms used in this text can be found in Appendix 1. Here, f ∈ R
denotes frequency in cycles/sec or Hz. In general the Fourier transform, X a ( f ), is complex.
Polar form As such, it can be expressed in polar form in terms of its magnitude Aa ( f ) = |X a ( f )| and
phase angle φa ( f ) = X a ( f ) as follows.

X a ( f ) = Aa ( f ) exp[ jφa ( f )] (1.2.17)

Magnitude, phase The real-valued function Aa ( f ) is called the magnitude spectrum of xa (t), while the real-
spectrum valued function φa ( f ) is called the phase spectrum of xa (t). More generally, X a ( f ) itself is
called the spectrum of xa (t). For a real xa (t), the magnitude spectrum is an even function of
f , and the phase spectrum is an odd function of f .
When a signal passes through a linear system, the shape of its spectrum changes. Systems
Filters designed to reshape the spectrum in a particular way are called filters. The effect that a linear
system has on the spectrum of the input signal can be characterized by the frequency response.

DEFINITION
Let S be a stable linear time-invariant continuous-time system with input xa (t) and output
1.5: Frequency Response
ya (t). Then the frequency response of the system S is denoted Ha ( f ) and defined
Ya ( f )
Ha ( f ) =
Xa( f )

Thus the frequency response of a linear system is just the Fourier transform of the output
divided by the Fourier transform of the input. Since Ha ( f ) is complex, it can be represented
by its magnitude Aa ( f ) = |Ha ( f )| and its phase angle φa ( f ) = Ha ( f ) as follows

Ha ( f ) = Aa ( f ) exp[ jφa ( f )] (1.2.18)

Magnitude, phase The function Aa ( f ) is called the magnitude response of the system, while φa ( f ) is called the
response phase response of the system. The magnitude response indicates how much each frequency
component of xa (t) is scaled as it passes through the system. That is, Aa ( f ) is the gain of
the system at frequency f . Similarly, the phase response indicates how much each frequency
component of xa (t) gets advanced in phase by the system. That is, φa ( f ) is the phase shift of
the system at frequency f . Therefore, if the input to the stable system is a pure sinusoidal tone
xa (t) = sin(2π F0 t), the steady-state output of the stable system is

ya (t) = Aa (F0 ) sin[2π F0 t + φa (F0 )] (1.2.19)

The magnitude response of a real system is an even function of f , while the phase response
is an odd function of f . This is similar to the magnitude and phase spectra of a real signal.
Indeed, there is a simple relationship between the frequency response of a system and the
spectrum of a signal. To see this, consider the impulse response.

20 Chapter 1 Signal Processing

DEFINITION Suppose the initial condition of a continuous-time system S is zero. Then the output of the
1.6: Impulse Response system corresponding to the unit impulse input is denoted h a (t) and is called the system
impulse response.
h a (t) = Sδa (t)

From the sifting property of the unit impulse in (1.2.13) one can show that the Fourier
transform of the unit impulse is simply a ( f ) = 1. It then follows from Definition 1.5 that
when the input is the unit impulse, the Fourier transform of the system output is Ya ( f ) = Ha ( f ).
That is, an alternative way to represent the frequency response is as the Fourier transform of
the impulse response.
Ha ( f ) = F{h a (t)} (1.2.20)
In view of (1.2.17), the magnitude response of a system is just the magnitude spectrum of
the impulse response, and the phase response is just the phase spectrum of the impulse response.
It is for this reason that the same symbol, Aa ( f ), is used to denote both the magnitude spectrum
of a signal and the magnitude response of a system. A similar remark holds for φa ( f ) which
is used to denote both the phase spectrum of a signal and the phase response of a system.

Example 1.4 Ideal Lowpass Filter

An important example of a continuous-time system is the ideal lowpass filter. An ideal lowpass
Ideal lowpass filter filter with cutoff frequency B Hz, is a system whose frequency response is the following pulse
of height one and radius B centered at f = 0.

1, | f | ≤ B
ρB ( f ) =
0, | f | > B
A plot of the ideal lowpass frequency response is shown in Figure 1.15.
Recall from Definition 1.5 that Ya ( f ) = Ha ( f )X a ( f ). Consequently, the filter in Fig-
ure 1.15 passes the frequency components of xa (t) in the range [−B, B] through the filter
without any distortion whatsoever, not even any phase shift. Furthermore, the remaining fre-
quency components of xa (t) outside the range [−B, B] are completely eliminated by the filter.
The idealized nature of the filter becomes apparent when we look at the impulse response of
the filter. To compute the impulse response from the frequency response we must apply the
inverse Fourier transform. Using the table of Fourier transform pairs in Appendix 1, this yields
h a (t) = 2B · sinc(2Bt)

FIGURE 1.15: Ha ( f )
Frequency
Response of Ideal 6
Lowpass Filter
1

- f
−B 0 B

1.3 Sampling of Continuous-time Signals 21

FIGURE 1.16: Impulse Impulse Response

Response of Ideal 250
Lowpass Filter
when B = 100 Hz
200

150

h (t)
100
a

−50
−0.04 −0.03 −0.02 −0.01 0 0.01 0.02 0.03 0.04
t (sec)

Here the normalized sinc function is defined as follows.

sin(π x)
sinc(x) =
πx
Sinc function The sinc function is a two-sided decaying sinusoid that is confined to the envelope 1/(π x). Thus
sinc(k) = 0 for k = 0. The value of sinc(x) at x = 0 is determined by applying L’Hospital’s
rule which yields sinc(0) = 1. Some authors define the sinc function as sinc(x) = sin(x)/x.
The impulse response of the ideal lowpass filter is sinc(2BT ) scaled by 2B. A plot of the
impulse response for the case B = 100 Hz is shown in Figure 1.16.

Notice that the sinc function, and therefore the impulse response, is not a causal signal.
But h a (t) is the filter output when a unit impulse input is applied at time t = 0. Consequently,
for the ideal filter we have a causal input producing a noncausal output. This is not possible
for a physical system. Therefore, the frequency response in Figure 1.15 cannot be realized
with physical hardware. In Section 1.4, we examine some lowpass filters that are physically
realizable that can be used to approximate the ideal frequency response characteristic.

• • • • • • • • • ••••• ••

1.3 Sampling of Continuous-time Signals

1.3.1 Sampling as Modulation
Periodic impulse train The process of sampling a continuous-time signal xa (t) to produce a discrete-time signal x(k)
can be viewed as a form of amplitude modulation. To see this, let δT (t) denote a periodic train
of impulses of period T .

∞
δT (t) = δa (t − kT ) (1.3.1)
k=−∞

22 Chapter 1 Signal Processing

FIGURE 1.17: δT (t)

Sampling as e
Amplitude
Modulation of an
Impulse Train ?
xa (t) e - × e x̂a (t)

Thus δT (t) consists of unit impulses at integer multiples of the sampling interval T . The
Sampled signal sampled version of signal xa (t) is denoted x̂a (t), and is defined as the following product.

x̂a (t) = xa (t)δT (t) (1.3.2)
Since x̂a (t) is obtained from xa (t) by multiplication by a periodic signal, this process is
Amplitude a form of amplitude modulation of δT (t). In this case δT (t) plays a role similar to the high-
modulation frequency carrier wave in AM radio, and xa (t) represents the low-frequency information signal.
A block diagram of the impulse model of sampling is shown in Figure 1.17.
Using the basic properties of the unit impulse in (1.2.11), the sampled version of xa (t) can
be written as follows.
x̂a (t) = xa (t)δT (t)
∞
= xa (t) δa (t − kT )
k=−∞

∞
= xa (t)δa (t − kT )
k=−∞
∞
= xa (kT )δa (t − kT ) (1.3.3)
k=−∞

Thus the sampled version of xa (t) is the following amplitude modulated impulse train.

∞
x̂a (t) = x(k)δa (t − kT ) (1.3.4)
k=−∞

Whereas δT (t) is a constant-amplitude or uniform train of impulses, x̂a (t) is a nonuniform

impulse train with the area of the kth impulse equal to sample x(k). A graph illustrating
the relationship between δT (t) and x̂a (t) for the case xa (t) = 10t exp(−t)u a (t) is shown in
Figure 1.18.
It is useful to note from (1.3.4) that x̂a (t) is actually a continuous-time signal, rather than
a discrete-time signal. However, it is a very special continuous-time signal in that it is zero
everywhere except at the samples where it has impulses whose areas correspond to the sample
values. Consequently, there is a simple one-to-one relationship between the continuous-time
signal x̂a (t) and the discrete-time signal x(k). If x̂a (t) is a causal continuous-time signal, we
Laplace transform can apply the Laplace transform to it. The Laplace transform of a causal continuous-time
signal xa (t) is denoted X a (s) and is defined
∞

X a (s) = L{xa (t)} = xa (t) exp(−st) dt (1.3.5)
0

It is assumed that the reader is familiar with the basics of the Laplace transform. Tables of
common Laplace transform pairs and Laplace transform properties can be found in Appendix 1.
Comparing (1.3.5) with (1.2.16) it is clear that for causal signals, the Fourier transform is just
the Laplace transform, but with the complex variable s replaced by j2π f . Consequently, the

1.3 Sampling of Continuous-time Signals 23

FIGURE 1.18: Periodic (a) Periodic Impulse Train

Impulse Train in (a) 2
and Sampled
Version of xa (t) in 1.5
(b) Using Impulse
Sampling 1

dT(t)
0.5

0
0 0.5 1 1.5 2 2.5 3 3.5 4
t (sec)
(b) Amplitude Modulated Impulse Train
4

3
xa(t)

0
0 0.5 1 1.5 2 2.5 3 3.5 4
t (sec)

spectrum of a causal signal can be obtained from its Laplace transform as follows.
X a ( f ) = X a (s)|s= j2π f (1.3.6)
At this point a brief comment about notation is in order. Note that the same base symbol, X a , is
being used to denote both the Laplace transform, X a (s), in (1.3.5), and the Fourier transform,
X a ( f ), in (1.2.16). Clearly, an alternative approach would be to introduce distinct symbols for
each. However, the need for additional symbols will arise repeatedly in subsequent chapters,
so using separate symbols in each case quickly leads to a proliferation of symbols that can
be confusing in its own right. Instead, the notational convention adopted here is to rely on
the argument type, a complex s or a real f , to distinguish between the two cases and dictate
the meaning of X a . The subscript a denotes a continuous-time or analog quantity. The less
cumbersome X , without a subscript, is reserved for discrete-time quantities introduced later.
If the periodic impulse train δT (t) is expanded into a complex Fourier series, the result can
be substituted into the definition of x̂a (t) in (1.3.2). Taking the Laplace transform of x̂a (t) and
converting the result using (1.3.6), we then arrive at the following expression for the spectrum
of the sampled version of xa (t).

∞
1
X̂ a ( f ) = X a ( f − if s ) (1.3.7)
T i=−∞

1.3.2 Aliasing
The representation of the spectrum of the sampled version of xa (t) depicted in (1.3.7) is called
Aliasing formula the aliasing formula. The aliasing formula holds the key to determining conditions under which
the samples x(k) contain all the information necessary to completely reconstruct or recover
xa (t) from the samples. To see this, we first consider the notion of a bandlimited signal.

24 Chapter 1 Signal Processing

Typically B is chosen to be as small as possible. Thus if xa (t) is bandlimited to B, then the

highest frequency component present in xa (t) is B Hz. It should be noted that some authors
use a slightly different definition of the term bandlimited by replacing the strict inequality in
Definition 1.7 with | f | ≥ B.
DEFINITION
A continuous-time signal xa (t) is bandlimited to bandwidth B if and only if its magnitude
1.7: Bandlimited Signal
spectrum satisfies
|X a ( f )| = 0 for |f| > B

The aliasing formula in (1.3.7) is quite revealing when it is applied to bandlimited signals.
Notice that the aliasing formula says that the spectrum of the sampled version of a signal is
just a sum of scaled and shifted spectra of the original signal with the replicated versions of
X a ( f ) centered at integer multiples of the sampling frequency f s . This is a characteristic of
amplitude modulation in general where the unshifted spectrum (i = 0) is called the base band
Base, side bands and the shifted spectra (i = 0) are called side bands. An illustration comparing the magnitude
spectra of xa (t) and x̂a (t) is shown in Figure 1.19.
Undersampling The case shown in Figure 1.19 corresponds to f s = 3B/2 and is referred to as undersam-
pling because f s ≤ 2B. The details of the shape of the even function |X a ( f )| within [−B, B]
are not important, so for convenience a triangular spectrum is used. Note how the sidebands in
Figure 1.19b overlap with each other and with the baseband. This overlap is an indication of
Aliasing an undesirable phenomenon called aliasing. As a consequence of the overlap, the shape of the
spectrum of x̂a (t) in [−B, B] has been altered and is different from the shape of the spectrum
of xa (t) in Figure 1.19a. The end result is that no amount of signal-independent filtering of
x̂a (t) will allow us to recover the spectrum of xa (t) from the spectrum of x̂a (t). That is, the
overlap or aliasing has caused the samples to be corrupted to the point that the original signal
xa (t) can no longer be recovered from the samples. Since xa (t) is bandlimited, it is evident

FIGURE 1.19: (a) Magnitude Spectrum of xa

Magnitude Spectra 2
of xa (t) in (a) and
x^a (t) in (b) when 1.5
B = 100, fs = 3B/2
|Xa(f)|

0.5

0
−300 −200 −fs −100 0 100 fs 200 300
f (Hz)
(b) Magnitude Spectrum of Sampled Signal

200
fd
150
|Xa(f)|

100

0
−300 −200 −fs −100 0 100 fs 200 300
f (Hz)

1.3 Sampling of Continuous-time Signals 25

that there will be no aliasing if the sampling rate is sufficiently high. This fundamental result
is summarized in the Shannon sampling theorem.

PROPOSITION
Suppose a continuous-time signal xa (t) is bandlimited to B Hz. Let x̂a (t) denote the
1.1: Signal Sampling
sampled version of xa (t) using impulse sampling with a sampling frequency of f s . Then the
samples x(k) contain all the information necessary to recover the original signal xa (t) if
f s > 2B

In view of the sampling theorem, it should be possible to reconstruct a continuous-time

signal from its samples if the signal is bandlimited and the sampling frequency exceeds twice
the bandwidth. When f s > 2B, the sidebands of X̂ a ( f ) do not overlap with each other
or the baseband. By properly filtering X̂ a ( f ) it should be possible to recover the baseband
and rescale it to produce X a ( f ). Before we consider how to do this, it is of interest to see
what happens in the time domain when aliasing occurs due to an inadequate sampling rate. If
aliasing occurs, it means that there is another lower-frequency signal that will produce identical
samples. Among all signals that generate a given set of samples, there is only one signal that
is bandlimited to less than half the sampling rate. All other signals that generate the same
Impostors samples are high-frequency impostors or aliases. The following example illustrates this point.

Example 1.5 Aliasing

The simplest example of a bandlimited signal is a pure sinusoidal tone that has all its power
concentrated at a single frequency F0 . For example, consider the following signal where
F0 = 90 Hz.
xa (t) = sin(180π t)

From the Fourier transform pair table in Appendix 1, the spectrum of xa (t) is
j[δ( f + 90) − δ( f − 90)]
Xa( f ) =
2
Thus xa (t) is a bandlimited signal with bandwidth B = 90 Hz. From the sampling theorem,
we need f s > 180 Hz to avoid aliasing. Suppose xa (t) is sampled at the rate f s = 100 Hz. In
this case T = .01 seconds, and the samples are

x(k) = xa (kT )
= sin(180π kT )
= sin(1.8π k)
= sin(2π k − .2π k)
= sin(2π k) cos(.2π k) − cos(2π k) sin(.2π k)
= − sin(.2π k)
= − sin(20π kT )

Thus the samples of the 90 Hz signal xa (t) = sin(180π t) are identical to the samples of the
following lower-frequency signal that has its power concentrated at 10 Hz.

xb (t) = − sin(20π t)

A plot comparing the two signals xa (t) and xb (t) and their shared samples is shown in
Figure 1.20.

Copyright 2010 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
Exploring the Variety of Random
Documents with Different Content
it be viewed in its place among facts of the same class. We propose,
therefore, without troubling the reader with details which are to be
found, at large, in many well-known works, and which he may be
supposed to have in recollection—or within his reach—to direct him
to a few principal points of the comparison which may be instituted
between the classical and the sacred writings, in relation to the
proof of the genuineness and authenticity of each kind.
The Jewish and Christian Scriptures may then be brought into
comparison with the works of the Greek and Roman authors, in the
following particulars:—
1. The number of manuscripts which passed down through the
middle ages, in the modes which have been described in the
preceding chapters.
About fifteen manuscripts of the history of Herodotus are known
to critics: and of these, several are not of higher antiquity than the
middle of the fifteenth century. One copy, in the French king’s library
(there are in that collection five or six), appears to belong to the
twelfth century; there is one in the Vatican, and one in the
Florentine library, attributed to the tenth century: one in the library
of Emmanuel College, Cambridge, formerly the property of
Archbishop Sancroft, which is believed to be very ancient: the
libraries of Oxford and of Vienna contain also manuscripts of this
author. This amount of copies may be taken as more than the
average number of ancient manuscripts of the classic authors; for
although a few have many more, many have fewer.
To mention any number as that of the existing ancient
manuscripts, either of the Hebrew or Greek Scriptures, would be
difficult. It may suffice to say that, on the revival of learning, copies
of the Scriptures, in whole or in part, were found wherever any
books had been preserved. In examining the catalogues of
conventual libraries—such as they were in the fifteenth century, the
larger proportion is usually found to consist of the works of the
fathers, or of the ecclesiastical writers of the middle ages; next in
amount are the Scriptures—sometimes entire; more often the
Gospels, the Acts, the Epistles, or the Psalms, separately; and last
and fewest are the classics, of which, seldom more than three or
four, are found in a list of one or two hundred volumes. The number
of ancient manuscripts of the Greek New Testament, or parts of it,
which hitherto have been examined by editors, is nearly five
hundred.
If in the case of a classic author, twenty manuscripts, or even five,
are deemed amply sufficient (and sometimes one, as we have seen,
is relied upon), it is evident that many hundreds are redundant for
the purposes of argument. The importance of so great a number of
copies consists in the amplitude of the means which are thereby
afforded of restoring the text to its pristine purity; for the various
readings collected from so many sources, if they do not always place
the true reading beyond doubt, afford an absolute security against
extensive corruptions.
2. The high antiquity of some existing manuscripts.
A Virgil (already mentioned) in the Vatican, claims an antiquity as
high as the fourth century: there are a few similar instances; but
generally the existing copies of the classics are attributed to periods
between the tenth and fifteenth centuries. In this respect the
Scriptures are by no means inferior to the classics. There are extant
copies of the Pentateuch, which, on no slight grounds, are supposed
to have been written in the second, or the third century: and there
are copies of the Gospels belonging to the third, or the fourth, and
several of the entire New Testament, which unquestionably were
made before the eighth. But the actual age of existing manuscripts is
a matter of more curiosity than importance; since proof of another
kind carries us with certainty some way beyond the date of any
existing parchments.
3. The extent of surface over which copies were diffused, at an
early date.
The works of the most celebrated of the Greek authors always
found a place in the libraries of opulent persons, in all parts of
Greece, and in many of the colonies, soon after their first
publication; and a century or two later they were read, wherever the
language was spoken. But a contraction of this sphere of diffusion
took place at the time when the eastern empire was being driven in
upon its centre; and during a long period these works were found
only in the countries and islands within a short distance of
Constantinople. As for the Latin classics, how widely soever they
might have been diffused during three or four centuries, the
incursions of the northern nations, and the consequent decline of
learning in the West, went near to produce their utter annihilation.
Many of these authors were actually lost sight of during several
centuries.
It is a matter of unquestioned history that the Jews, always
carrying with them their books, had spread themselves throughout
most countries of Asia, of southern Europe, and of northern Africa,
before the commencement of the Christian era; nor is it less certain
that, wherever Judaism existed, Christianity rapidly followed it.
Carried forward by their own zeal, or driven on by persecutions, the
Christian teachers of the first and second centuries passed beyond
the limits of the Roman empire, and founded churches among
nations that were scarcely known to the masters of the world. Nor
were the Christian Scriptures merely carried to great distances in
different directions;—they were scattered through the mass of
society, in every nation, to an extent greatly exceeding the ordinary
circulation of books in those ages: these books were not in the
hands of the opulent, and of the studious merely; for they were
possessed by innumerable individuals, who, with an ardour beyond
the range of secular motives, valued, preserved, and reproduced
them. And while many copies were hoarded and hidden by private
persons, others were the property of societies, and, by continual
repetition in public, the contents of them were imprinted on the
memories of their members.
The wide, and—if the expression may be used—the deep
circulation of the Scriptures, preserved them, not merely from
extinction, but, to a great extent, from corruption also. These books
were at no time included within the sphere of any one centre of
power—civil or ecclesiastical. They were secreted, and they were
expanded far beyond the utmost reach of tyranny or of fraud.
4. The importance attached to the books by their possessors.
In a certain sense, the religion of the Greeks and Romans was
embodied in the works of their poets; but the religious fervour of the
people had never linked itself with those works, as if they were the
depositories of their faith: books were the possession of the opulent
and the educated classes;—they were prized by the few as the
means of intellectual enjoyment. But Judaism first, and Christianity
not less, were religions of historical fact: the doctrines and the laws
of these religions were inferences, arising naturally from the belief of
certain memorable events, and from the expectation of other events,
that were yet to take place; the record of the past had become at
once the rule of duty, and the charter of hope. To the dispersed and
hated Jew his books were the solace of his wounded national pride:
to the persecuted Christian his books were his title to “a better
country,” and his support under present privations and sufferings. If
the canonical books are valued by the Christian of modern times
who believes them to be divine, they were valued with a far deeper
sense by the early Christians, who, on the ground of undoubted
miracles, received them as the word of Him who is omnipotent.
The veneration felt by the Jews for their sacred books was of a
kind that is altogether without parallel: the reverence of the
Christians for theirs, if it was not more profound, was much more
impassioned, and this feeling gave intensity to a sentiment wholly
unlike any with which one might seek to compare it: the fondness of
a learned Greek or Roman for his books, was but in comparison as
the delight of a child with his toys.
To this deep feeling towards the sacred writings, in the minds of
Christians, was owing, not only the concealment and the
preservation of copies in times of active persecution; but the
assiduous reproduction of them by persons of all ranks who found
leisure to occupy themselves in a work which they deemed to be so
meritorious, and which they found to be so consoling.
5. The respect paid to them by copyists of later ages.
We have seen that, throughout the middle ages, though nothing
like a widely diffused taste for the classic authors existed, yet at all
times, there were, here and there, individuals by whom they were
read and valued, and by whose agency and influence so much care
was bestowed upon their preservation as served to insure a safe
transmission of them to modern times. But that the Latin authors, at
any time after the decline of the western empire, received the
benefit of a careful and competent collation of copies there is little
reason to believe. Of the Greek authors there were issued new
recensions from Alexandria, while that city continued to be the seat
of learning; and some measure of the same care was exercised by
the scholars of Constantinople; yet even there the celebrated works
of antiquity suffered a great degree of neglect during the last four
centuries of the eastern empire.
But in this respect, as well as in those already mentioned, the text
of the Scriptures—Jewish and Christian—possesses an incomparable
advantage over that of the classic authors. The scrupulosity and the
servile minuteness of the Jewish copyists in transcribing the Hebrew
Scriptures are well known; in a literal sense of the phrase, “not a
tittle of the law” was slighted: not only—as with the Greeks—was the
number of verses in each book noted, but the number of words and
of letters; and the central letter of each book being distinguished, it
became, as a point of calculation, the key-stone of that portion of
the volume. This unexampled exactness affords security enough for
the safe transmission of the text; and if there were any grounds for
the suspicion that the Rabbis, to weaken the evidence adduced
against them by the Christians, wilfully corrupted some particular
passages, we have other security, as we shall see, against the
consequences of such an attempt.
The flame of true piety was, at no time, extinguished in the
Christian community; nor can any century or half century of the
middle ages be named, in relation to which it may not be proved
that there were individuals by whom the books of the New
Testament were known and regarded with a heartfelt reverence and
affection. There were, besides, multitudes in the religious houses
who, influenced perhaps by superstitious notions, thought it a work
of superlative merit to execute a fair copy of the Scriptures, or any
part of them; and all the adornments which the arts of the times
afforded, were lavished to express the veneration of the scribe for
the subject of his labours.
And more than this;—the Scriptures, especially in the first eight
centuries, underwent several careful and skilful revisions in the
hands of learned and able men, who, collating all the copies they
could procure, restored the text wherever, as they thought, errors
had been admitted. The prodigious labours of Origen in restoring the
text of the Septuagint version have been often spoken of. The
fathers of the Western, the African, and the Asiatic Churches—
especially Jerome, Eusebius, and Augustine, with such means as
they severally possessed, did what they could to stop the progress of
accidental corruption in the sacred text, by instituting new
comparisons of existing copies.
6. The wide local separation, or the open hostility of those in
whose custody these books were preserved.
This is a circumstance of the utmost significance, and if it be not
peculiar to the Jewish and Christian Scriptures, yet it belongs to
them in a degree which places their uncorrupted preservation on a
basis immeasurably more extended and substantial than that of any
other ancient writings. The Latin authors were scantily dispersed
over the Roman world, and never were they in the keeping of distant
nations, or hostile parties. The Greek classics had indeed, to some
extent, come into the hands of the western nations, as well as of the
Greeks, in earlier times, and during the middle ages. And, if any
weight can be attached to the fact, some of these works were also
in the keeping of the Arabians: but they were never the subject of
mutual appeal by rival communities.
The Hebrew nation has, almost throughout the entire period of its
history, been divided, both by local separation, and by schisms.
Probably the Israelites of India, and certainly the Samaritans, have
been the keepers of the books of Moses—apart from the Jews,
during a period that reaches beyond the date of authentic profane
history. Throughout times somewhat less remote the Jews have not
only been separated by distance, but divided by at least one
complete schism—that on the subject of the Rabbinical traditions,
which has distinguished the sect of the Karaites from the mass of
the nation.
The reproach of the Christian Church—its sects and divisions—has
been, in part at least, redeemed by the security thence arising, for
the uncorrupted transmission of its records. Almost the earliest of
the Christian apologists avail themselves of this argument in proof of
the integrity of the sacred text. Augustine especially urged it against
those who endeavoured to impeach its authority: nor was there ever
a time when an attempt, on any extensive scale—even if otherwise it
might have been practicable—to alter the text would not have raised
an outcry in some quarter. From the earliest times the common Rule
of Faith was held up for the purposes of defence or of aggression by
the Church, and by some dissentient party. Afterwards the partition
of the Christian community into two hostile bodies, of which Rome
and Constantinople were the heads, afforded security against any
general consent to effect alterations of the text. And in still later
ages a few uncorrupted communities, existing within the bounds of
the Romish Church, became the guardians of the sacred volume.
7. The visible effects of these books from age to age.
On this point also the history of the Greek and Latin classics
affords only the faintest semblance of that evidence by means of
which the existence and influence of the Scriptures may be traced
from the earliest times after their publication, through all successive
ages. The Greek and Latin authors indicated their continued
existence scarcely at all beyond the walls of schools and halls of
learning. During a full thousand years the world saw them not—
governments did not embody them in their laws or institutions;—the
people had no consciousness of them. They were less known, and
less thought of abroad, than were the ashes of the dead—than the
bones, teeth, blood, tears, and tatters of the Greek and Romish
martyrs.
How different are the facts that present themselves on the side of
the Jewish and Christian Scriptures! The Jews—in the sight of all
nations—have, through a well-known and uncontested period of two
thousand five hundred years, exhibited a living model of the
venerable volume which was so long ago delivered to them, and
which still they fondly cherish. And though long since debarred from
the enjoyment of all that was splendid or cheering in their
institutions, and though rent away from their land as well as their
worship, and though too often blind to the moral grandeur of their
law, and mistaken in the meaning of their prophets, they hold
unbroken the shell of the religious system which is described in their
books. Whatever in their religion was of less value—whatever served
only to cover and protect the vital parts—whatever was the most
peculiar, and the least important, whatever might have been laid
aside without damage or essential change, has been retained by
these wanderers; while that which was precious—the sacred books
excepted—has been lost.
The Christian Scriptures have marked their path through the field
of time, not in the regions of religion only, or of learning, or of
politics; but in the entire condition—moral, intellectual, and political
—of the European nations. The history of no period since the first
publication of these writings can be intelligible apart from the
supposition of their existence and diffusion. If we look back along
the eighteen centuries past, we watch the progress of an influence,
sometimes indeed marking its presence in streams of blood—
sometimes in fires, sometimes by the fall of idol temples, sometimes
by the rearing of edifices decked with new symbols; nor can the
distant and mighty movement be explained otherwise than by
knowing that the books we now hold and venerate were then
achieving the overthrow of the old and obstinate evils of idolatry. It
is needless to say that the history of Europe in all subsequent
periods has implied, by a thousand forms of false profession, and by
the constancy of the few, the continued existence of the Christian
Scriptures.
8. The body of references and quotations.
The successive references of the Greek authors, one to another,
though they are amply sufficient, in most instances, to establish the
antiquity of the works quoted, furnish a very imperfect aid in
ascertaining the purity of the existing text, or in amending it where
apparently it is faulty. A large number of these references are merely
allusive, consisting only of the mention of an author’s name, with
some vague citation of his meaning. And even in those authors who
make copious and verbal quotations, such as Strabo, Plutarch,
Hesychius, Aulus Gellius, Stobæus, Marcellinus, Photius, Suidas, and
Eustathius, a lax method of quotation, in many instances, robs such
quotations of much of their value for purposes of criticism. And yet,
after every deduction of this kind has been made, the reader of the
classics feels an irresistible conviction that this network of mutual or
successive references could not have resulted from machination,
contrivance, or from anything but reality; it affords a proof, never to
be refuted, of the genuineness of the great mass of ancient
literature.
But as to the Jewish and Christian Scriptures, this kind of
evidence, reaching far beyond the mere proof of antiquity and
genuineness, is ample and precise enough to establish the integrity
of nearly the entire text of the books in question. These writings
were not simply succeeded by a literature of a similar cast; but they
actually created a vast body of literature altogether devoted to their
elucidation; and this elucidation took every imaginable form of
occasional comment upon single passages—of argument upon
certain topics, requiring numerous scattered quotations, and of
complete annotation, in which nearly the whole of the original
author is repeated. From the Rabbinical paraphrases, and from the
works of the Christian writers of the first seven centuries (to come
later is unnecessary) the whole text of the Scriptures might have
been recovered if the originals had since perished.
If any one is so uninformed as to suppose that this kind of
evidence is open to uncertainty, or that it admits of refutation, let
him, if he has access to an ordinary English library, open the
volumes of writers of all classes since the days of Elizabeth, and see
how many allusions to Shakespeare, and how many verbal
quotations from his plays, and how many commentaries upon
portions, or upon the whole of them he can find; and then let him
ask himself if there remains the possibility of doubting that these
dramas—such in the main as they now are, were in existence at the
accession of James I. If these quotations and allusions were not
more than a fifth or a tenth part of what they actually are, the proof
would not, in fact, be less conclusive than it is.
9. Early versions.
For the purpose of establishing the antiquity, genuineness, and
integrity of the Scriptures, no other proof need be adduced than that
which is afforded by the ancient versions now extant. When
accordant translations of the same writings, in several unconnected
languages, and in languages which have long ceased to be
vernacular, are in existence, every other kind of evidence may be
regarded as superfluous.
In this respect a comparison between the classic authors and the
Scriptures can barely be instituted; for scarcely anything that
deserves to be called a translation of those writers—executed at a
very early period after their first publication, is extant. But, on the
other side, the high importance attached by the Jews to the Old
Testament, and by the early Christians to the New, and the earnest
desire of the poor and unlearned to possess, in their own tongue,
the words of eternal life, suggested the idea, and introduced the
practice, of making complete and faithful translations of both.
Thus it is that, independently of the original text, the Old
Testament exists in the Chaldee paraphrases or Targums; in the
Septuagint, or Greek version; in the translations of Aquila, of
Symmachus, and of Theodosian; in the Syriac and the Latin, or
Vulgate versions; in the Arabic, and in the Ethiopic; not to mention
others of later date.
The New Testament has been conveyed to modern times, in whole
or in part, in the Peshito, or Syriac translation, in the Coptic, in
several Arabic versions, in the Æthiopic, the Armenian, the Persian,
the Gothic, and in the old Latin versions.
10. The vernacular extinction of the languages, or of the idioms, in
which these books were written.
To write Attic Greek was the ambition and the affectation of the
Constantinopolitan writers of the third and fourth centuries; and thus
also, to acquire a pure Latinity, was assiduously aimed at by writers
of the middle ages; and, in fact, a few of them so far succeeded in
this sort of imitation that they executed some forgeries, on a small
scale, which would hardly have been detected, if they had not
wanted external proof.
But now the pure Hebrew—such as it had been spoken and
written before the Babylonish captivity, had so entirely ceased to be
vernacular during the removal of the Jews from their land, that
immediately after their return the original Scriptures needed to be
interpreted to the people by their Rabbis; nor is there any evidence
that the power of writing the primitive language was affected by
these Rabbis, whose commentaries are composed in the dialect that
was vernacular in their times.
As to the Hellenistic Greek of the New Testament, differing as it
does, from the style of the classic authors, and even from that of the
Septuagint, to which it is the most nearly allied, it very soon passed
out of use; for the later Christian writers, in the Greek language,
had, in most instances, formed their style before the time of their
conversion; or at least they aimed at a style, widely differing from
that of the apostles and evangelists. The idiom of the New
Testament, in which phrases or forms of speech borrowed from the
surrounding languages occur, resulted from the very peculiar
education and circumstances of the writers, which were such as to
make their dialect, in many particulars, unlike any other style; and
such as could not fail soon to become extinct.
11. The means of comparison with spurious works; or with works
intended to share the reputation that had been acquired by others.
Imitations—whether good or bad—are useful in serving to set
originals in a more advantageous light. Good imitations, calling into
activity, as they do, all the acumen and the utmost diligence of
critics, enable them to place genuine writings out of the reach of
suspicion. Bad imitations, by serving as a foil or contrast, exhibit
more satisfactorily, the dignity, the consistency, and the simplicity of
what is genuine.
Several good imitations of the style of Cicero have appeared in
different ages, and they have called for so much acuteness on the
part of critics as have materially strengthened the evidence of the
genuineness of his acknowledged works. In like manner the
celebrated epistles of Phalaris excited a learned and active
controversy, the beneficial result of which was not so much the
settling of the particular question in debate, as the concentration of
powerful and accomplished minds upon the general subject of the
genuineness of ancient books, by means of which other questionable
remains of antiquity received the implicit sanction of retaining their
claims, after they had been brought within the reach of so fiery an
ordeal.
Many bad imitations of classic authors have been offered to the
world, and some such are still extant; and sometimes these are
appended to the author’s genuine works. No one can read these
spurious pieces immediately after he has made himself familiar with
such as are genuine, without receiving, from the contrast, a forcible
impression of the truth and reality of the latter. The life of Homer, for
example, which is usually appended to the history of Herodotus, and
which claims his name, and which has something of his manner, yet
presents a contrast which few readers can fail to observe.
No good imitations, either of the Jewish or Christian Scriptures,
have ever appeared; but in the place of that elaborate investigation
which the existence of such productions would have called forth,
other motives of the strongest kind have prompted a fuller and more
laborious examination of the Scriptures than any other writings have
endured.
Bad imitations of the style of the Scriptures—some of the Old
Testament, and many of the New—have been attempted, and are
still in existence; and they are such as to afford the most striking
illustration that can be imagined of the difference in simplicity,
dignity, and consistency, which one should expect to find, severally,
in the genuine and the spurious. The apocryphal books (which
however are not, most of them, properly termed spurious) afford an
advantageous contrast in this way, to the genuine or canonical
writings of the Old Testament; and as to the spurious gospels—
passing under the names of Peter, Judas, Nicodemus, Thomas,
Barnabas—a very cursory examination of them is enough to
enhance, immeasurably, the confidence we feel in the genuineness
of the true Gospels and Epistles.
The preservation of these latter worthless productions to modern
times, is an extraordinary fact, and it affords proof of a state of
things, the knowledge of which is important in questions of literary
antiquity—namely, that there were many copyists in the middle ages
who wrote, and went on writing, mechanically, whatever came in
their way, without exercising any discrimination. Now there is more
satisfaction in knowing that ancient books have come down through
a blind and unthinking medium of this sort, than there would be in
believing that we possess only such things as the copyists, in the
exercise of an assumed censorship, deemed worthy to be handed
down to posterity. It is far better that we should—by accident and
ignorance, have lost some valuable works, and that, by the same
means, some worthless ones have been preserved, than that the
results of accident and ignorance should have been excluded by the
constant exercise of a power of selection governed by, we know not
what rule or influence. Nothing more pernicious can be imagined
than the existence, from age to age, of a synod of copyists sagely
determining what works should be perpetuated, and what should be
suffered to fall into oblivion. Happily for literature and religion, there
were, in the monasteries, numbers of unthinking labourers, who, in
selecting the subject of their mindless toils, seemed to have followed
the easy rule of taking—the next book on the shelf!
12. The strength of the inference that may be drawn from the
genuineness of the books to the credibility of their contents.
Nothing can be more simple or certain than the inference derived
from the acknowledged antiquity and genuineness of an historical
work, in proof of the general credibility of the narrative it contains. If
it be proved that Cicero’s Orations against Catiline, and that Sallust’s
History of the Catiline War, were written by the persons whose
names they bear; or if it were only proved that these compositions
were extant and well known as early as the age of Augustus; that
they were then universally attributed to those authors, and were
universally admitted to be authentic records of matters of fact; and if
the same facts are, with more or less explicitness, alluded to by the
writers of the same, and of the following age, there remains no
reasonable supposition, except that of the truth of the story—in its
principal circumstances, by aid of which the existence and the
acceptance of these narratives, these orations, and these allusions,
so near to the time of the conspiracy, can be accounted for.
In Sallust’s History some facts may be erroneously stated; or the
principal facts may be represented under the colouring of prejudice.
In the Orations of Cicero there may be (or we might for argument
sake suppose there to be) exaggeration, and an undue severity of
censure; but after any such deductions have been made, or any
others which reason will allow, it remains incontestably certain that,
if these writings be genuine, the story, in the main, is true. The
sophisms of a college of sceptics, in labouring to show the
improbability of the facts, or the suspiciousness of the evidence,
would not avail to shake our belief if we are convinced that the
books are not spurious.
Nor is this inference less direct, or less valid in the case above
mentioned, than in any similar instance of more recent occurrence.
It is as inevitable to believe that Catiline conspired against the
Roman state, and failed in the attempt, as that the descendants of
James II. excited rebellions in Scotland, or that a French General
was for a short time king of Naples. In the one case, as in the
others, unless the documents—all of them, have been forged, the
facts must be true.
The principle upon which such an inference is founded, scarcely
admits of an exception. Narratives of alleged, but unreal facts, may
have been suddenly promulgated, and for a moment credited; or
false narratives of events—concealed by place or circumstances from
the public eye, may have gained temporary credit. Or narratives,
true in their outline, may have been falsified in all those points
concerning which the public could not fairly judge; and thus the
false, having been slipped in along with the true, has passed, by
oversight, upon the general faith. But no such suppositions meet the
case of various public transactions, taking place through some length
of time, and in different localities, and which were witnessed by
persons of all classes, interests, and dispositions, and which were
uncontradicted by any parties at the time, and which were
particularly recorded, and incidentally alluded to, by several writers
whose works were widely circulated—generally accepted, and
unanswered, in the age when thousands of persons were competent
to judge of their truth.
No one—to recur to the example mentioned above, is at liberty
merely to say that he withholds his faith from Sallust, and from
Cicero, as he might, on many points, withhold it from Herodotus,
from Diodorus, or from Plutarch. Yet even in that case, he ought to
show cause of doubt, if he would not be charged with the frivolous
affectation of possessing more sagacity than his neighbours pretend
to. But in the other case, while in professing to doubt the facts, he is
not able to impugn the antiquity of the records, he only gives
evidence of some want of coherence in his modes of thinking. He
who professes not to believe the narrative, should be required to
give an intelligible account of the existence of the writings, on the
supposition that the events never took place.
When historical facts which, in their nature, are fairly open to
direct proof, are called in question, it is an irksome species of trifling
to make a halt upon twenty indirect arguments, while the centre
proof—that which a clear mind fastens upon intuitively, remains
undisposed of. In an investigation that is purely historical, and which
is as simple as any that the page of history presents, it boots
nothing to say that the books of the New Testament contain
doctrines which do not accord with our notions of the great system
of things; or that they enjoin duties which are grievous and
impracticable; or that they favour despotism, or engender strifes. It
avails nothing to say that some professors of Christianity are
hypocrites, and that therefore the religion is not true. No objections
of this sort weaken in any way that evidence upon which we believe
that our island was once possessed by the Romans. But yet they
have as much weight in counterpoising that evidence, as they have
in balancing the proof of the facts that are affirmed in the New
Testament. If such objections were ten-fold more valid than
sophistry can make them, they would not remove, or alter, or impair,
one grain of the proper proof, belonging to the historical proposition
under inquiry.
The question is not whether we admire and approve of
Christianity, or not; or whether we wish to submit our conduct to its
precepts, and to abide by the hope it offers; or intend to risk the
hazards of it being true. The question is not whether, in our opinion,
these books have been a blessing to the world, or the contrary; but
simply this—whether the religion was promulgated and its
documents were extant, and were well known throughout the
Roman empire, in the reign of Nero.
There are evasions enough, by means of which we may remove
from our view the inference which follows from an admission of the
antiquity and genuineness of the Christian Scriptures. But
contradiction may be challenged when it is affirmed that, if the
Gospels, the Acts of the Apostles, the Epistles of Paul, of Peter, of
John, and of James, were written in the age claimed for them, and
were immediately diffused throughout Palestine, Asia Minor, Africa,
Greece, and Italy, then this fact carries with it inevitably the truth of
the Christian system.
Remote historical facts, though incapable of that kind of palpable
proof which overrules contradiction, are yet open to a kind of proof
which no one who really understands it can doubt. Just on this
ground stand all the main facts of ancient history;—they are
inevitably admitted as true by all into whose minds the whole of the
evidence enters; and they are believed or doubted, in every degree
between blind faith and blind scepticism, by those whose
apprehension of the facts is defective, or obscure, or perverted.
When it is said that the events recorded in the four Gospels are
presented to us in a form that has been purposely adapted to
exercise our faith, it should be added, by way of illustrating the
exact meaning of the words—that the events recorded by
Thucydides and Tacitus are also presented to us in a form that is
adapted to exercise our faith. Yet it would be more exactly proper to
say—that this sort of evidence is adapted to give exercise to reason;
for faith has no part in things which come within the known
boundaries of the system in the midst of which we are called to act
our parts. And here it should be understood that facts (intelligible in
themselves) may, in the fullest sense, be supernatural, and yet when
they are duly attested, in conformity with the ordinary principles of
evidence, they as much belong to the system with which we are
every day concerned, as do the most familiar transactions of
common life.
The Scriptures do indeed make a demand upon our faith; but this
is exclusively in relation to facts which belong to a world above and
beyond that with which we are conversant, and of which facts we
could know nothing by any ordinary means of information. Our
assent to miraculous events, when properly attested, is demanded
on the ground of common sense: the facts themselves are as
comprehensible as the most ordinary occurrences; and the evidence
upon which they are attested implies nothing beyond the well-known
principles of human nature. If then we reject this evidence, we
exhibit, not a want of faith, for that is not called for; but a want of
reason. To one who affected to question the received account of the
death of Julius Cæsar, we should not say “you want faith,” but “you
want sense.” It is the very nature of a miracle to appeal to the
evidence of universal experience, in order that, afterwards, a
demand may be made upon faith, in relation to extra-mundane
facts.
CHAPTER XV.
ILLUSTRATIONS OF THE PRECEDING
STATEMENTS:—A MORNING AT THE BRITISH
MUSEUM.

And now, at this stage of our progress, let the reader indulge the
author to the extent of a page of metaphor or allegory.—Imagine
then that we are standing on the margin of a mighty river, the
opposite shore of which is scarcely visible; and as to the origin of
this world of waters, it is far remote and is unknown:—as to the
ocean into which it shall at length empty itself and its treasures—this
is distant also, nor do we find it anywhere laid down in our maps.
The flow of this river is tranquil—its surface is glassy; but upon this
surface there float samples and fragments, innumerable, of the
products of each of the countries which it has watered in its course:
—here come rafts, laden with well-packed bales, and there,
confusedly mingled, are things more than can be counted—torn
away—rent—shattered—coated with rust—wrapped around with
weeds. Moving onward, we see the symbols and the devices of
nations long ago extinct, and the utensils of a forgotten civilisation,
and the products of lands—thousands of miles up the stream; and
these entangled with the symbols, the devices, the rare and curious
products, of some country next above us. On the bosom of this
mighty river there float samples of all things, and these commingled
in all imaginable modes.
This is our day-dream:—now for the interpretation of it. We have
imagined ourselves to be stationed in any one of the saloons of the
British Museum; or that we are passing up and down, from one of
these halls to another: and at length are coming to a rest in the
centre of the New Reading-room. The countless collections of
antiquities—marbles—coins—gems—utensils—weapons—costumes—
the manuscripts—the illuminations, and the printed books—what are
all these things, but so many relics of remote ages which, favoured
by various chances, have floated down to this, our own era, upon
the broad surface of the River of Time?
But are these tens of thousands—these hundreds of thousands of
individual objects, are they so many disjointed and disconnected
particles?—this is far from being the fact. It is a very small number
of things, in this vast collection, concerning which an instructed
Curator would acknowledge his ignorance, as to what it is, and to
what age it belongs, and of what country or people it is a relic. As to
a thousand to one of all the single contents of the British Museum,
each of them links itself, either nearly, or remotely, with the nine
hundred, ninety and nine, of its neighbours—right and left; or
perhaps with some articles that are exhibited in the opposite wing of
the building: as for instance—here is a coin, the legend upon which
we should have failed to read, or to understand, had it not been that
a Greek writer, of whose works a sole manuscript has come down to
modern times, incidentally mentions a fact concerning some obscure
town of Asia Minor, and its history, under the Roman emperors, of
which otherwise we should have been ignorant.
Let us avail ourselves of another supposition, remote as it may be
from the fact; and it is this—That the author, and the reader, of this
book, whom we imagine to be now pacing together the saloons of
the Museum, are possessed of that universality of learning, and that
vastness of antiquarian accomplishment, which enables the
gentleman at the centre table of the Reading-room to answer all
inquirers, and to aid and guide them all in carrying forward their
various researches. If, then, the author and the reader were gifted in
any such manner as this, we might then, with a sort of second sight,
or a veritable clairvoyance, look upon the countless stores around us
as if they were all falling into an appointed order, or were obeying
some natural law of mutual attraction and cohesion: as thus—there
goes an almost illegible manuscript, attaching itself to a colossal
sculpture—much as feathers stick themselves on to an electric
conductor:—there are coins, arranging themselves spontaneously,
like a crown of laurel leaves, around the brows of busts:—there are
weapons and fragments of armour, edging themselves on to a copy
of Polybius:—there are bits of a pediment, or the chippings of a
column, claiming a standing-place upon the Greek text of Procopius
—and why? it is because these fragments belong to an edifice of the
times of Justinian, which he has described. And now, as to the
printed books, and the manuscripts, whence many of the printed
books drew their existence, if we will give way to the ideal for a few
moments, we shall see them floating out from their shelves, in this
vast circus, and knowingly arranging themselves, in a sort of
pyramidal form, as if to exhibit their real relationship of quotation,
and of reference, in the order of time—the more recent to the more
ancient—the many to the few;—until the pile—made up of a million
of books, is surmounted by the two or three that quote none older
than themselves, and that are quoted by all.
What then is our inference? It is this: that as to the persons and
the events—the doings and the notions—the thoughts and the ways
—the customs and the manners—the philosophy—the literature—the
religion—the politics—the civilisation, of the nations of all those ages
which are comprehended within the limits of what is called the
historic period—these innumerable matters are assuredly, known to
us, at this time;—and they have become known to us with this
degree of certainty (in the main) not by the precarious and insulated
testimony of a few writers, whose works have reached modern times
—we know not how; but very much otherwise than thus; for it is by
means of the inter-related, and the mutually attestative evidence of
thousands of witnesses—witnesses in stone and marble, in metallic
substances, coins and brass plates, in membranous records, and in
writings upon every other material, and in every imaginable fashion;
and all these things are so netted together and so welded, and
dove-tailed, and linked, and glued, and sealed, into a vast
conglomerate, as that the combined testimony thence accruing in
support of our voluminous historic beliefs is not less solid than are
the granitic ribs of a continent; and is as various, and as rich, as all
the products of its surface—its faunas and its floras.
So much for a momentary glance at the treasures, the vast
accumulations of the British Museum;—but now we might usefully
take the Synopsis in hand, and give attention to some few of the
articles that are named in it. What we are in search of are those
attestations of ancient written evidences, touching the persons, the
events, the manners, the religions, of ancient nations, which come
upon us—we might say, by surprise, and which are derived from
sources altogether and in every sense independent, and
unconnected, one with another.
Take with you, in one hand, your Tacitus, Sallust, Dion Cassius;—
and in the other hand, your Virgil, Horace, Juvenal, and Ovid. These
writers—the one set as historians, the other set as poets, build up to
our view the throne, and its personages, of the Imperial Times—say,
of two centuries, reckoned back from the life-time of the last of
them. But through what channels have the books come into our
hands? The editors of the printed copies assure us that there had
come into their possession, in each instance, one, two, three
manuscripts, that had been raked out of the forgotten heaps of this
or that monastery, or other conservatory of curious articles. As to
the greater number of these manuscripts, they could not be
assigned to an age much beyond the ninth century; therefore, on
the supposition that they are genuine works—the products of a time
seven hundred, or a thousand years earlier, what the editor had
under his eye must have been nothing better than a copy—from a
copy—or perhaps, from several in succession! Is not this line of
proof somewhat precarious? Ought we to trust ourselves to it?
Advance toward the left hand, from the entrance hall, and by the
time you have moved on a dozen steps, the volumes in your hands,
if they were gifted with consciousness, would begin to twitch and to
jerk themselves about, as if uneasy in being held away from their old
friends, right and left, whom they recognise, perched on the
pedestals, and fixed to the walls. Whence is it that these solid
antiquities have been brought hither? Not from those same lumber-
vaults in the monasteries, or the royal libraries of Europe, whence
we have received the aforesaid manuscripts;—not so, but from deep
under-ground—from cavities—from underneath pavements, sixty
feet or more lower than the present surface: they have been picked
up in cornfields; they have been sifted from out of heaps of rubbish;
they have been taken from the recesses of the houses of a city,
buried by a volcanic eruption, many centuries ago. These manifold
samples of an ancient civilisation have been fished up from the beds
of rivers and the bottoms of lakes; and these recoveries have been
effected in all these and many other modes over the extent of
Europe, and of Southern and Western Asia, and of North Africa.
There is no possibility therefore of calling in question this million-
tongued testimony; we must not gainsay what is affirmed by these
tongues of stone and of brass, of silver and of gold.
And the more, in any instance, the coincidence is slender and
remote, or, as one might say, frivolous or unimportant, so much the
surer, and the more to be relied upon is it, in what it does affirm: as
thus—Look to your Synopsis, page 87, Compartment III.:—“A pig of
lead, inscribed with the name of the Emperor Domitian, when he
was consul for the eighth year, A.D. 82, weighing 154 lbs. It was
discovered in 1731 underground, on Hayshaw Moor, in the West
Riding of Yorkshire, half-way between an ancient lead-mine, north of
Pately Bridge, and the Roman road from Ilkley (Olicana) to
Aldborough (Isurium).” This pig had slept where he was dropped
about 1,650 years.
“A pig of lead, inscribed with the name of the Emperor Hadrian,
weighing 191 lbs.; found in 1796 or ’97, at Snailbeach Farm, parish
of Westbury, 10 miles south-west of Shrewsbury.” Then follow some
other pigs, whose slumbers underground have been more or less
prolonged and profound.
“A pig of lead, inscribed with the name of L. Aruconius
Verecundus, and the letters Metal. Lvtvd., probably the mine of
Lutudæ. Found near Matlock Bank, in Derbyshire.”
“A pig of lead, inscribed CL. TR. LVT. BR. EX. ARG.; found with
three other pigs and some broken Roman pottery, at Broomer’s Hill,
in the parish of Pulborough, Sussex, January 31, 1824, close to the
Roman Road, Stone Street, from London to Chichester.”
“A pig of lead, inscribed with the name of Britannicus, the son of
the Emperor Claudius; found on the Mendip Hills, Somersetshire.”
So much for these pigs. What is it which they might say, if we
were to bring them into court? Something of this sort: At this time,
in the streets of the stannary towns in Cornwall, there are to be seen
blocks—pigs of tin, stamped in a manner similar to the lettering of
these pigs of lead in the British Museum. This stamping is effected
for the purpose of securing the dues of the Duchy of Cornwall, and
the symbols and the letters indicate the political fact that the Prince
of Wales, as Duke of Cornwall, lays a hand upon every pound of tin
that is smelted in the county; and thus, too, the stamping of the
produce of the lead-mines of Britain gives evidence of the fact that
the Romans were not merely resident in Britain at the time, but were
masters also of the island, and the lords of its mineral products.
Then the lettering itself finds its interpretation in the Roman imperial
history, and this history comes into our hands, partly as it has been
narrated by the Roman historians above mentioned; partly in the
form of sculptures, statues, busts, and bas-reliefs; and partly, and
very copiously, in the unquestionable form of the coins of the same
emperors, which alone would suffice for putting us in possession of
the series of events, greater and smaller, through a course of many
centuries. But what the reader should here keep in view is this: that
as our present thesis is—the safe and sure transmission of ancient
books, by the means of often-repeated copyings, through the lapse
of ages, an evidence to this effect—and it is the most conclusive that
can be imagined or desired, is afforded us when, in passing through
collections, such as those treasured in the British Museum, the Books
in question are found to furnish a coherent, and a continuous, and
an exact interpretation of these palpable and ponderous antiquities.
Yet, it is manifest that, unless the books were in the main genuine,
they could not have supplied any interpretations, such as are those
which we find in them.
Go on now to the historical sculptures—the statues, and the busts
of the imperial times. These, for the most part, are susceptible of
authentication by means of the coins of the same emperors, which
may be seen—by “order”—in another department of the Museum;
the likenesses are indisputable, and the historic reality of the two
samples of Roman art is thus far made good. But beyond this we
may safely go. From the Roman writers—specially Tacitus,
Suetonius, and Dion Cassius—we acquire what we need not doubt to
be a true idea of the individual character, the temperament, the
education, the public and private behaviour, and the style of the
series of imperial persons, from Julius Cæsar, onward, to the times
of each of these writers. What then is the verdict of our
physiognomical instincts, when we compare the busts or statues, for
instance, of Augustus and of Tiberius, of Nero and of Trajan? We
could no more take these, one for the other, than we could misname
the portraits of Philip of Spain, or the Duke of Alva, put by the side
of George Washington, or John Howard; or misjudge those of Oliver
Cromwell, and John Milton; or of Admiral Blake, and Alexander Pope.
We need not wait until a science of physiognomy has been
concocted before we may risk a guess in writing the names under
portraits of Lord Chatham, Dr. Johnson, and Oliver Goldsmith.
Mistakes, in single instances, may be made, but not in the long run;
and when, on the one hand, we take the entire series of royal
portraits, eastern and western, from the first of the Ptolemies to
Charlemagne, and, on the other hand, the books of the series of
contemporary historians, we shall receive, from this large collation of
independent evidences, an irresistible conviction of the general
authenticity of the latter; and therefore we must cease to entertain
doubts on this question of the secure transmission of ancient books
to modern times.
It would be of little avail here to cite a few single instances of the
agreement of Roman coins with written history, for such instances
are countless. The reader who would wish to inform himself, in
whole, or in part, on this extensive subject, should take in hand a
Medallic History of Imperial Rome, which, as compared with the
medallic treasures of the British Museum, will give him aid in
following the train of public events through five or six centuries,
exhibited and verified by the double line of testimonies—the metallic
and the literary. Or he may be content to take, as a sufficient sample
of this species of proof, the facts he will find brought together in a
small volume, “Akerman on the Coins of the Romans relating to
Britain.”
There is another field upon which a gleaning, and more than a
gleaning, may easily be made by help of the Roman poets as our
guides. These writers—and we need name only Ovid, Horace, Virgil,
and Propertius—are undoubtedly believed to have lived and
flourished as the contemporaries of Julius Cæsar, Augustus, Tiberius.
Their writings, as we have them now in our hands, are accepted as
genuine; for the criticism which demonstrates the general integrity
of the text (exceptions allowed for) is too erudite and careful to be
called in question. Consequently, these writings have safely
traversed a period of fifteen hundred years, ending with the date of
the earliest printed editions: but this transit has been made by no
other means than that of the copyists; and therefore, if, as we shall
see, a super-abundance of various and independent evidences
removes the possibility of our doubting the fact, then this mode of
transmission, precarious as it may seem, is found to be trustworthy,
and our main point is established—namely, That ancient books have
indeed come down to modern times—whole and entire. Let us look,
for a moment, to this corroborative evidence—such as we find it
offered to the eye, in passing through the saloons of the British
Museum.
The Roman poets were not, perhaps, themselves very firm
believers in the Grecian mythology—considered religiously or
historically: nevertheless, they took it up—such as it had come into
their hands—and it was a splendid inheritance—a boundless treasury
of bright conceptions of superhuman power, beauty, grace; a
scheme of elegant sensuousness, with a touch of sublimity. Its
fables, far more available for poetic purposes than any system of
serious truths could have been, opened before the Roman poets a
broad meadow land, in roaming through which the imitative, more
than originative turn of the Roman mind, might gather fruits and
flowers, ripe and gay, and which asked only to be taken and
enjoyed. So it is, therefore, that in every imaginable mode of
lengthened poetic narrative, and of transient allusion, and of direct
and of allusive reference, the gods and the goddesses, and the
demi-gods, and the heroes of Greece come up upon the stage of the
Roman poetry. These repetitions—these borrowings or plagiarisms,
and these flashing glances, are countless:—sometimes they are
formal; sometimes they are informal:—they are broad daylight views
in some places, and in places innumerable they are as sparks only—
visible for an instant.
Now with what objects is it that these mythologic passages are in
harmony?—with what is it that they correspond? Our answer is—
With tens of thousands of relics of ancient art which, through
channels altogether independent of those through which the books
have reached us, have come, at this time, to fill, and to over-fill the
cabinets and museums of Europe—and thus, also, our British
Museum.
But then this mass of ponderable and visible evidences is inter-
related in a very peculiar manner, which should be borne in mind.
We have just now referred to the correspondence which connects
the historic sculptures—the statues and the busts of Roman
personages, male and female, and the likenesses of the same men
and women which are so copiously supplied in collections of the
Roman imperial mintage. But now we pass on to the Græco-Roman
saloons—the first, the second, and the third, as well as the
basement-room. These are filled with mythologic sculptures—
recovered from the soil of Italy and Greece:—they show us, in
inimitable marbles, those same divinities, the principal and the
subordinate, which the mind of Greece had imagined, and which the
Roman artists adopted: these beautiful creations we at once
recognise as the celestial personæ with whom we have made
acquaintance in the pages of the Roman poets: the conception of
superhuman grace and power is the very same; and the attendant
symbols are the same. And now furnish yourself with the requisite
order for inspecting the collection of antique gems—precious (often)
as to their material—precious, incalculably more so, by means of
that exquisite taste and that inimitable executive skill which have
made them what they are.
These microscopic sculptures, in consequence of the value of the
material, and the costliness of the work, and from their smallness,
and the facility of preservation, were eagerly sought after by the
opulent at the very time of their production; and they have been
most carefully hoarded in every age, by the same class of persons;
and they have suffered far less injury in the lapse of time than
antiquities of any other kind. Especially the intaglios—the indented
sculptures, are, for the most part, as perfect and sharp now as they
were eighteen hundred years ago. What is it, then, that these gems
of art bring under our modern eyes?—it is the very same ideal
personages of the same mythology;—and the symbols are the same,
and the air, and the grace, and the attributes of beauty and power
are the same;—there is the same sensuousness—there are the same
ambiguous adventures;—there is the same poetry and the same art
—poetry and art, admirable, indeed, how much soever it may be
open to censure as to its moral quality.
Here then we have in view three independent, but perfectly
concurrent and mutually interpretative evidences—namely, first, the
sculptures, secondly, the gems, and then the books—the poetry. If,
in examining one of these classes of antiquities, we find ourselves at
a loss in attempting to decipher its symbols or its allusions, any such
difficulty vanishes—in most instances—when we betake ourselves to
another class:—as thus—the gem expounds the statue; or the poet,
in a single verse, sheds his beam of light upon both. Thus it is that—
with the three at our command—antiquity, throughout the rich and
splendid region of its mythologies, stands unveiled before us! Must
we not grant that so many coherences, and so many
correspondences, and so many interpretative agreements—countless
as they are—can have had their source in nothing but the realities of
the age whence we believe them to have descended to modern
times? But if it be so, then it is true that ancient books—to wit, the
Roman poets—have been securely sent forward—thanks to the
copyists!—from age to age, through all the intervening years of so
many centuries.
If it were a volume that was now to be filled, instead of the few
pages of this chapter, and if, instead of a morning at the British
Museum, an entire season were to be diligently spent there, we
should still want space and leisure for specifying a sample only of
those articles which might properly be referred to in illustration of
our present argument. Instead of doing so, we must move forward
through the Elgin Saloon, only stopping to make this one observation
—that these sculptures, and these bas-reliefs, and these inscriptions,
would be to us, at this time, nothing better than a vast confusion—a
mass of insoluble enigmas, if we did not carry with us the written
remains of the Greek and Roman literature—the works of the
historians, and the poets, and the dramatists, and the orators, which
were the creations of that same age of refined intelligence, and
exquisite taste, and artistic skill: but so it is, that the written
memorials of that brief period are found to be available for
interpreting the solid memorials of the same times, and these again
for illustrating those. It was indeed a brief period: —it was a
blossoming and a fruit-bearing summer month of the world’s dull
millennial year; and during the long period that followed it—the
autumn months, and the winter—there were none among the living
who could either have written these books, or who could have
chiselled these marbles; but the books in one manner, and the
marbles in another, have separately floated down upon the billows of
time; and here we have them, confronted under one roof—ten
thousand witnesses, attesting the reality of ancient history.
From the classic antiquities we now advance, and enter the
Assyrian Galleries. Everybody knows, or may easily know, in what
way the sculptures, buried so many centuries, have now come to fill
these long apartments, and how they thus find a resting-place under
the roof of the British Museum. The places whence they have come,
and the circumstances of their disinterment, are (as we must
suppose) known and familiar to the visitor in whose company we are
spending this morning in its saloons. This being so, and if, moreover,
we may believe that he has become, in some degree, conversant
with the literature of ancient Greece—especially with its historians—
Herodotus, Thucydides, Xenophon, Diodorus, and also Strabo—he
will be qualified to understand what we mean in speaking of that
broad confirmation of the authenticity of ancient history which it
receives from a glance at the contents of the “Assyrian Galleries.”
The above-named Greek writers, and these illustrated as they are
by the contemporary literature, give us a distinct image of Greece,
and of its people, with their intellectuality, and their religion, and
their taste; and this portraiture is quite homogeneous in itself, and,
as we have already said, it is corroborated and exhibited, in ten
thousand instances, by the sculptures, and other objects found in
the saloons we have just now visited. But now these same writers
open up to us also—sometimes formally, and sometimes incidentally
—a prospect, eastward, far over the regions outstretched beyond the
limits of the Greek civilisation. In those illimitable expanses there
existed a civilisation; but it was quite of another aspect; there was
government, and social order; but these were wholly unlike the
institutions of Greece. There were religions; but they breathed
another spirit: they uttered other voices; they spoke of a different
national economy. There was the same human nature; but it had
been developed as if under conditions proper to another world.
Now I will ask my companion to tell me with what sort of feeling it
is, that, in passing from the monuments of Grecian life, and the
remains of its arts, he enters these Assyrian galleries. Does there not
take place an involuntary impression to this effect—as if we were
here setting foot upon the soil of another world? We have crossed
the threshold that divides one phase or mode of human existence
from another mode of it; there are here displayed before us the
indications of a different climate, a different terrestrial surface; and
the vegetation that covers it is of another class, nor are the animals

As NZS 3679.1 (2016)
100% (1)
As NZS 3679.1 (2016)
57 pages
SAP Note 2351294 - S4HANA System Conversion - Upgrade - Measures To Reduce Technical Downtime 2351294 - E - 20221221
No ratings yet
SAP Note 2351294 - S4HANA System Conversion - Upgrade - Measures To Reduce Technical Downtime 2351294 - E - 20221221
10 pages
Diy Drones For The Evil Genius: Design, Build, and Customize Your Own Drones: Design, Build, and Customize Your Own Drones 1St Edition - Ebook PDF
100% (7)
Diy Drones For The Evil Genius: Design, Build, and Customize Your Own Drones: Design, Build, and Customize Your Own Drones 1St Edition - Ebook PDF
51 pages
Just Walk-Out Technology and Its Challenges:A Case of Amazon Go
No ratings yet
Just Walk-Out Technology and Its Challenges:A Case of Amazon Go
4 pages
Marsis Ingest User Manual v152
No ratings yet
Marsis Ingest User Manual v152
16 pages
A State of Art Techniques On Machine Learning Algorithms A Perspective of Supervised Learning Approaches in Data Classification
100% (1)
A State of Art Techniques On Machine Learning Algorithms A Perspective of Supervised Learning Approaches in Data Classification
5 pages
IIB (v9 & v10) Continuous Integration-Maven-Jenkins
100% (1)
IIB (v9 & v10) Continuous Integration-Maven-Jenkins
19 pages
Baofeng Uv5r Chirp Programming Guide
100% (1)
Baofeng Uv5r Chirp Programming Guide
3 pages
Galxe Protocol Draft
No ratings yet
Galxe Protocol Draft
19 pages
Calibre Realtime Integration To Cadence Virtuoso - 2017.1: Initial Setup and Invocation Session Window Integrated Toolbar
No ratings yet
Calibre Realtime Integration To Cadence Virtuoso - 2017.1: Initial Setup and Invocation Session Window Integrated Toolbar
2 pages
Isocon en Is Rev01
No ratings yet
Isocon en Is Rev01
14 pages
BRKSPG 2557
No ratings yet
BRKSPG 2557
55 pages
Samsung Foundry Ln04lpp s00 Memory Compiler Rf2 Hs v1.52 Datasheet Rev1.52
No ratings yet
Samsung Foundry Ln04lpp s00 Memory Compiler Rf2 Hs v1.52 Datasheet Rev1.52
49 pages
Soft, Hard and Ruby Hard Real Time With Linux
No ratings yet
Soft, Hard and Ruby Hard Real Time With Linux
37 pages
Software Cadmatic Saw Control System
No ratings yet
Software Cadmatic Saw Control System
19 pages
An Introduction To Websim A Briefing Guide
No ratings yet
An Introduction To Websim A Briefing Guide
26 pages
Complete Download Windows Internals Part 2 Developer Reference 7th Edition Russinovich PDF All Chapters
No ratings yet
Complete Download Windows Internals Part 2 Developer Reference 7th Edition Russinovich PDF All Chapters
52 pages
Aruba Clearpass Network Admission Control (NAC Solution) + Wireless Solution Wifi 6E - 3 Years
No ratings yet
Aruba Clearpass Network Admission Control (NAC Solution) + Wireless Solution Wifi 6E - 3 Years
10 pages
Instant download Theory of Agglomerative Hierarchical Clustering Sadaaki Miyamoto pdf all chapter
100% (3)
Instant download Theory of Agglomerative Hierarchical Clustering Sadaaki Miyamoto pdf all chapter
65 pages
International Catalogue 2019
No ratings yet
International Catalogue 2019
47 pages
Virtualization Project Proposal
No ratings yet
Virtualization Project Proposal
8 pages
1 NGBSS Solution Overview
No ratings yet
1 NGBSS Solution Overview
34 pages
Bravo User Manual PDF
No ratings yet
Bravo User Manual PDF
18 pages
OTDR Certification Brochure - 11410-00820 Rev. B
No ratings yet
OTDR Certification Brochure - 11410-00820 Rev. B
2 pages
Videoedge-camera-handler-V50100 RN Rm0 LT en
100% (1)
Videoedge-camera-handler-V50100 RN Rm0 LT en
275 pages
C++ in Embedded Systems - Myth and Reality
No ratings yet
C++ in Embedded Systems - Myth and Reality
9 pages
Scalping EA - GOLD 30M Back Test Report
No ratings yet
Scalping EA - GOLD 30M Back Test Report
2 pages
Money Management - XLSX - MASTER
No ratings yet
Money Management - XLSX - MASTER
1 page
03 - BVMS 10.0 BVMS Installation - PGV28
No ratings yet
03 - BVMS 10.0 BVMS Installation - PGV28
21 pages
Assignment
No ratings yet
Assignment
29 pages
Quanta QLM - DA0QLMMB8E0 - LG Raffaello2 - Rev1A - LG A530 - Schematic
No ratings yet
Quanta QLM - DA0QLMMB8E0 - LG Raffaello2 - Rev1A - LG A530 - Schematic
42 pages
VLLM: Using PagedAttention To Optimize LLM Inference and Serving
No ratings yet
VLLM: Using PagedAttention To Optimize LLM Inference and Serving
6 pages
Ixchariotusersguide 710 SP 4
No ratings yet
Ixchariotusersguide 710 SP 4
474 pages
Csta v5 Users Guide-Ed1
No ratings yet
Csta v5 Users Guide-Ed1
75 pages
SC-STC-001 - Module 1 - Introduction & Security Desk
No ratings yet
SC-STC-001 - Module 1 - Introduction & Security Desk
21 pages
Gtu Mpi Paper Solution
No ratings yet
Gtu Mpi Paper Solution
19 pages
15h00 OMMIC - F LECOURT PDF
No ratings yet
15h00 OMMIC - F LECOURT PDF
45 pages
Anurag Resume
No ratings yet
Anurag Resume
2 pages
2024 CFA Program Curriculum Level II Volume 4 Equity Valuation Fixed Income 1st Edition Cfa Institute all chapter instant download
100% (3)
2024 CFA Program Curriculum Level II Volume 4 Equity Valuation Fixed Income 1st Edition Cfa Institute all chapter instant download
65 pages
Chapter Three Software Anaylys
100% (2)
Chapter Three Software Anaylys
23 pages
Jito Validator Presentation
No ratings yet
Jito Validator Presentation
17 pages
Stock Price Pattern Recognition - A Recurrent NN Approach - K Kamijo T Tanigawa
0% (1)
Stock Price Pattern Recognition - A Recurrent NN Approach - K Kamijo T Tanigawa
7 pages
Client Server Distributed Computing
0% (1)
Client Server Distributed Computing
6 pages
ThousandEyes - Tests
No ratings yet
ThousandEyes - Tests
12 pages
SCC2560 Solution Selling - April 2022
No ratings yet
SCC2560 Solution Selling - April 2022
82 pages
Case Study
No ratings yet
Case Study
14 pages
SECSB - 001 - EN-2001 - Arbor DDoS Attack Protection Solutions - 0
No ratings yet
SECSB - 001 - EN-2001 - Arbor DDoS Attack Protection Solutions - 0
2 pages
ERP Sandip V1
0% (1)
ERP Sandip V1
4 pages
Oracle Cloud SCM Assessment Answers
No ratings yet
Oracle Cloud SCM Assessment Answers
5 pages
TLB 100 Line Cross PDF
No ratings yet
TLB 100 Line Cross PDF
17 pages
William Stallings Computer Organization and Architecture 8 Edition Reduced Instruction Set Computers
No ratings yet
William Stallings Computer Organization and Architecture 8 Edition Reduced Instruction Set Computers
23 pages
Wago Remote Io With Fanuc Robot Using Ethernet/Ip Protocol: A500390, English
No ratings yet
Wago Remote Io With Fanuc Robot Using Ethernet/Ip Protocol: A500390, English
14 pages
التحليل الفني للاسواق المالية عبد المجيد المهيلمي 3
No ratings yet
التحليل الفني للاسواق المالية عبد المجيد المهيلمي 3
334 pages
AEM 3.3 Administration Guide (00073UM)
No ratings yet
AEM 3.3 Administration Guide (00073UM)
79 pages
Project
No ratings yet
Project
13 pages
MVC Framework Tutorial
100% (1)
MVC Framework Tutorial
90 pages
Toshiba GRR100
No ratings yet
Toshiba GRR100
16 pages
Virtualizing Business Critical Applications
100% (1)
Virtualizing Business Critical Applications
4 pages
Fundamentals of Digital Signal Processing Using MATLAB 2nd Edition Robert J. Schilling - Get the ebook instantly with just one click
100% (1)
Fundamentals of Digital Signal Processing Using MATLAB 2nd Edition Robert J. Schilling - Get the ebook instantly with just one click
47 pages
4708356
No ratings yet
4708356
81 pages
Fundamentals of Digital Signal Processing Using MATLAB 2nd Edition Robert J. Schilling 2025 scribd download
100% (2)
Fundamentals of Digital Signal Processing Using MATLAB 2nd Edition Robert J. Schilling 2025 scribd download
79 pages
Download full (Ebook) Fundamentals of Digital Signal Processing Using MATLAB by Robert J. Schilling, Sandra L Harris ISBN 9780840069092, 084006909X ebook all chapters
No ratings yet
Download full (Ebook) Fundamentals of Digital Signal Processing Using MATLAB by Robert J. Schilling, Sandra L Harris ISBN 9780840069092, 084006909X ebook all chapters
82 pages
Complete Download Problems Confronting Contemporary Democracies Essays in Honor of Alfred Stepan 1st Edition Douglas Chalmers PDF All Chapters
100% (7)
Complete Download Problems Confronting Contemporary Democracies Essays in Honor of Alfred Stepan 1st Edition Douglas Chalmers PDF All Chapters
71 pages
Instant download The Art of Hokusai 1st Edition Heather Rodino pdf all chapter
100% (7)
Instant download The Art of Hokusai 1st Edition Heather Rodino pdf all chapter
71 pages
Complete Download Teachers Parents And Classroom Behaviour 1st Edition Miller PDF All Chapters
100% (7)
Complete Download Teachers Parents And Classroom Behaviour 1st Edition Miller PDF All Chapters
81 pages
Instant Access to Asymmetric Trade Negotiations 1st Edition Sanoussi Bilal ebook Full Chapters
100% (6)
Instant Access to Asymmetric Trade Negotiations 1st Edition Sanoussi Bilal ebook Full Chapters
81 pages
201.26-RP1 5-08 YCWL0056 Through YCWL0610 R-410A PDF
No ratings yet
201.26-RP1 5-08 YCWL0056 Through YCWL0610 R-410A PDF
72 pages
Hospital Diets
100% (1)
Hospital Diets
451 pages
(TAI8540D) Humidity Sensor Module
No ratings yet
(TAI8540D) Humidity Sensor Module
3 pages
TD-12 MIDI Implementation: 1. Receive Data
No ratings yet
TD-12 MIDI Implementation: 1. Receive Data
14 pages
Linear Equations in Two Variables
No ratings yet
Linear Equations in Two Variables
24 pages
IBADAN MEDICAL FITNESS
No ratings yet
IBADAN MEDICAL FITNESS
1 page
SEATWORK-LBM 1st2324 STUDENTS
No ratings yet
SEATWORK-LBM 1st2324 STUDENTS
3 pages
LAPP MainCatalogue 1617 en Lowres
50% (2)
LAPP MainCatalogue 1617 en Lowres
1,223 pages
Sustainable Fashion Practices in India
No ratings yet
Sustainable Fashion Practices in India
8 pages
Beamer PDF
No ratings yet
Beamer PDF
40 pages
AISC Channel Design 360-16
No ratings yet
AISC Channel Design 360-16
6 pages
Assignment 1 HRM
No ratings yet
Assignment 1 HRM
8 pages
OOPS & DS Assignment Questions
No ratings yet
OOPS & DS Assignment Questions
2 pages
Career Guide 2020-Final PDF
No ratings yet
Career Guide 2020-Final PDF
851 pages
6 Heat of Combust and Phase W Picture
No ratings yet
6 Heat of Combust and Phase W Picture
4 pages
Download Full Learning Apache Kafka, Second Edition Nishant Garg PDF All Chapters
100% (1)
Download Full Learning Apache Kafka, Second Edition Nishant Garg PDF All Chapters
55 pages
Dave Pelz's Short Game Bible (PDFDrive)
No ratings yet
Dave Pelz's Short Game Bible (PDFDrive)
508 pages
Jahnvi Patel
No ratings yet
Jahnvi Patel
85 pages
Szalai Nóra - 5 Perc Angol Nyelvtan - Igeidők - Megoldókulcs
No ratings yet
Szalai Nóra - 5 Perc Angol Nyelvtan - Igeidők - Megoldókulcs
21 pages
Mgt201 Final Term Solved Papers Vuzs Team
100% (1)
Mgt201 Final Term Solved Papers Vuzs Team
7 pages
How To Start A Hotel Business
No ratings yet
How To Start A Hotel Business
25 pages
(eBook PDF) Introduction to Psychology 9th Edition pdf download
100% (5)
(eBook PDF) Introduction to Psychology 9th Edition pdf download
43 pages
Theory and Experiment On Microstrip Antennas
No ratings yet
Theory and Experiment On Microstrip Antennas
9 pages
General Obligations Under FIU-InD Guidelines_Krypsm Private Limited
No ratings yet
General Obligations Under FIU-InD Guidelines_Krypsm Private Limited
22 pages
ROAD
No ratings yet
ROAD
1 page
WinCC Programming en-US en-US
No ratings yet
WinCC Programming en-US en-US
2,688 pages
Test 1
No ratings yet
Test 1
40 pages
Overview of Hydropower Potential and Energy Sector in Myanmar
100% (1)
Overview of Hydropower Potential and Energy Sector in Myanmar
27 pages