Explore the full ebook collection and download it now at textbookfull.

com

Conceptual Digital Signal Processing with MATLAB

Keonwook Kim

https://textbookfull.com/product/conceptual-digital-signal-
processing-with-matlab-keonwook-kim/

OR CLICK HERE

DOWLOAD EBOOK

Browse and Get More Ebook Downloads Instantly at https://textbookfull.com

Click here to visit textbookfull.com and download textbook now
 Your digital treasures (PDF, ePub, MOBI) await
Download instantly and pick your perfect format...

Read anywhere, anytime, on any device!

Understanding Digital Signal Processing with MATLAB® and

Solutions. First Edition. Edition Poularikas

https://textbookfull.com/product/understanding-digital-signal-
processing-with-matlab-and-solutions-first-edition-edition-poularikas/

textbookfull.com

Digital Signal Processing-A Primer With MATLAB® 1st

Edition Samir I. Abood (Author)

https://textbookfull.com/product/digital-signal-processing-a-primer-
with-matlab-1st-edition-samir-i-abood-author/

textbookfull.com

Real Time Digital Signal Processing from MATLAB to C with

the TMS320C6x DSPs Third Edition Morrow

https://textbookfull.com/product/real-time-digital-signal-processing-
from-matlab-to-c-with-the-tms320c6x-dsps-third-edition-morrow/

textbookfull.com

MATLAB Signal Processing Toolbox User s Guide The

Mathworks

https://textbookfull.com/product/matlab-signal-processing-toolbox-
user-s-guide-the-mathworks/

textbookfull.com
Digital Signal Processing with Matlab Examples Volume 3
Model Based Actions and Sparse Representation 1st Edition
Jose Maria Giron-Sierra (Auth.)
https://textbookfull.com/product/digital-signal-processing-with-
matlab-examples-volume-3-model-based-actions-and-sparse-
representation-1st-edition-jose-maria-giron-sierra-auth/
textbookfull.com

Digital Image Processing A Signal Processing and

Algorithmic Approach 1st Edition D. Sundararajan (Auth.)

https://textbookfull.com/product/digital-image-processing-a-signal-
processing-and-algorithmic-approach-1st-edition-d-sundararajan-auth/

textbookfull.com

Smartphone based real time digital signal processing Third

Edition Abhishek Sehgal

https://textbookfull.com/product/smartphone-based-real-time-digital-
signal-processing-third-edition-abhishek-sehgal/

textbookfull.com

From Algebraic Structures to Tensors Digital Signal and

Image Processing Matrices and Tensors in Signal Processing
Set 1st Edition Gérard Favier (Editor)
https://textbookfull.com/product/from-algebraic-structures-to-tensors-
digital-signal-and-image-processing-matrices-and-tensors-in-signal-
processing-set-1st-edition-gerard-favier-editor/
textbookfull.com

Audio And Speech Processing With MATLAB Paul R. Hill

https://textbookfull.com/product/audio-and-speech-processing-with-
matlab-paul-r-hill/

textbookfull.com
Signals and Communication Technology

Keonwook Kim

Conceptual
Digital Signal
Processing
with MATLAB
Signals and Communication Technology

Series Editors
Emre Celebi, Department of Computer Science, University of Central Arkansas,
Conway, AR, USA
Jingdong Chen, Northwestern Polytechnical University, Xi'an, China
E. S. Gopi, Department of Electronics and Communication Engineering, National
Institute of Technology, Tiruchirappalli, Tamil Nadu, India
Amy Neustein, Linguistic Technology Systems, Fort Lee, NJ, USA
H. Vincent Poor, Department of Electrical Engineering, Princeton University,
Princeton, NJ, USA
This series is devoted to fundamentals and applications of modern methods of
signal processing and cutting-edge communication technologies. The main topics
are information and signal theory, acoustical signal processing, image processing
and multimedia systems, mobile and wireless communications, and computer and
communication networks. Volumes in the series address researchers in academia
and industrial R&D departments. The series is application-oriented. The level of
presentation of each individual volume, however, depends on the subject and can
range from practical to scientific.
**Indexing: All books in “Signals and Communication Technology” are indexed
by Scopus and zbMATH**
For general information about this book series, comments or suggestions, please
contact Mary James at mary.james@springer.com or Ramesh Nath Premnath at
ramesh.premnath@springer.com.

More information about this series at http://www.springer.com/series/4748

Keonwook Kim

Conceptual Digital Signal

Processing with MATLAB

123
Keonwook Kim
Division of Electronics
and Electrical Engineering
Dongguk University
Seoul, Korea (Republic of)

ISSN 1860-4862 ISSN 1860-4870 (electronic)

Signals and Communication Technology
ISBN 978-981-15-2583-4 ISBN 978-981-15-2584-1 (eBook)
https://doi.org/10.1007/978-981-15-2584-1
© Springer Nature Singapore Pte Ltd. 2021
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, expressed or implied, with respect to the material contained
herein or for any errors or omissions that may have been made. The publisher remains neutral with regard
to jurisdictional claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,
Singapore
To my parents
To Saea, Saeyun, and my lovely wife
Gumhong Kim
To all of my family
Preface

Digital signal processing (DSP) is an extensive study area to understand the signal
and noise in the digital domain. The advanced digital technology in the 60s and 70s
started to accelerate the theoretical signal processing for the feasible engineering
over the academia. In the 80s, the research in articles were organized and consol-
idated as practical textbooks for the graduate program. There were great textbooks
such as “Discrete-Time Signal Processing” by Oppenheim and Schafer and “Digital
Filter Design” by Parks and Burrus. The decade of the 90s further enjoyed the
processor technologies along with programming environments for DSP. MATLAB
(founded in 1984) has provided a reliable and convenient computer language (also
tools) to realize and verify the mathematical algorithms. Many textbooks appreciate
the benefits of MATLAB, for instance, “Signals and Systems Using MATLAB” by
Chaparro. For DSP, books and tools are existing.
This book addresses DSP in novel approach based on reconfiguration. The
intuitions and theories along with applications are the basic philosophy of the book.
The author tries to avoid the dictionary style (bottom-up structure) and to place
theories within applications. Students often experience a loss of motivation in the
middle of a semester due to the fragmented knowledge of digital filter design. The
author believes that the great researchers in DSP did not establish solid theories
from pure mathematics. The imagination initiates the research, and the equation
finalizes the theory. This book tries to provide the theories with derivations,
illustrations, and/or applications. Once the student has a certain picture of the
procedure and intelligence, the knowledge can be extended and maintained for a
long time. MATLAB also helps to design the top-down structure for the
self-motivated learning process. The book is organized as follows.

Chapter 1
• Basics of digital filters based on the intuition,
• Simple weight and sum of recent inputs for desired output,
• Design the low, high, and band pass filter.

vii
viii Preface

Chapter 2
• Definition of frequency in continuous and discrete time domain,
• Sampling theory,
• Discrete time signal representations.
Chapter 3
• Fundamentals to find the frequency magnitude from time domain,
• Discrete-time Fourier transform (DTFT) for non-periodic signal,
• Discrete Fourier transform (DTF) for periodic signal.
Chapter 4
• Reason to have the linear time invariance property,
• Finite impulse response (FIR) filter,
• Simple FIR filter design from the specification.
Chapter 5
• Extension of DTFT and DFT to Z-transform,
• Z-transform for infinite impulse response (IIR) filter,
• Intuitive IIR filter designs.
Chapter 6
• Understanding the filter specification,
• Advanced FIR filter designs,
• Advanced IIR filter designs.
Chapter 7
• Quantization effect for fixed-point number system,
• Implementation matters for FIR and IIR filters,
• Frequency domain filter realization.
Chapter 8
• Various filter realizations in MATLAB,
• Fixed-point number for MATLAB,
• C code generation for MATLAB.
In addition to above, there are two more sections for MATALB fundamentals
and Symbolic Math Toolbox as an appendix. This book includes comprehensive
information from basic DSP theories to the real-time filter realization of digital
computers and processors. The DSP class from this book is intended to provide
one- or two-semester program at the undergraduate level. The instructor can design
the DSP class for theory or practical intensive program as follows. Also, a
two-semester class can be handled using the complete contents of this book.
Preface ix

One-semester DSP program (theory intensive)

• Chapters 1+2+3+4+5
• Chapter 6
– Specification + FIR window method + FIR types,
– IIR Butterworth method with bilinear transform.
One-semester DSP program (practical intensive)
• Chapters 1+2+3+4+5
• Chapter 6 (Specification + IIR Butterworth method with bilinear transform),
• Chapter 7 (numerical representation + filter implementation components),
• Chapter 8 (whole).
Two-semester DSP program
• First semester: Theory-intensive DSP program as shown above.
• Second semester: Chapter 5 to end.
DSP is realized by programming languages which can be chosen from the
rudimentary level such as assembly language to the advanced level, for example,
MATLAB. Unlike the low-level language, MATLAB provides a high degree of
freedom for syntax, variable, function, execution, etc. Also, the robust support for
symbolic mathematics can solve the calculus problems without using the hand
derivations. The conversion between the symbolic and numerical mathematics is
seamless in MATLAB; hence, the equation establishment is the only requirement
for DSP realization, in most of the situations. Examples in this book demonstrate
the visualization, verification, and realization of DSP algorithms based on the
MATLAB programming. The evolution of DSP is continuously happening to the
next-generation applications such as data science, deep learning, etc. The author of
this book hopes that the readers can grasp the fundamentals of DSP and employ the
understandings of DSP to real-world applications.

Seoul, Korea (Republic of) Keonwook Kim

Contents

1 Preliminary Digital Filter Design . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.1 Discrete Time Signal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Design the Digital Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Filter Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.4 Digital Filter Definitions and Requirements . . . . . . . . . . . . . . . . . 12
1.5 What We Need for Further (Optional) . . . . . . . . . . . . . . . . . . . . . 16
1.6 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2 Frequency and Signals in Discrete Domain . . . . . . . . . . . . . . . . . . . . 21
2.1 Continuous Sinusoid Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.2 Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
2.3 Signal Representation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
2.4 Frequency in Discrete Time Signal . . . . . . . . . . . . . . . . . . . . . . . 56
2.5 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
3 Fourier Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
3.1 How We Can Find the Magnitude of the Specific Frequency? . . . . 71
3.2 Signal Property and Frequency Distribution . . . . . . . . . . . . . . . . . 81
3.3 Discrete-Time Fourier Transform (DTFT) . . . . . . . . . . . . . . . . . . 83
3.4 Discrete Fourier Transform (DFT) . . . . . . . . . . . . . . . . . . . . . . . . 89
3.5 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
4 Filters in Time Domain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
4.1 Digital Filter Revisited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
4.2 Filter Properties with LTI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
4.3 FIR and IIR Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
4.4 Frequency Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136

xi
xii Contents

4.5 Simple Filter Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142

4.6 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
5 Z-Transform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
5.1 Definition and Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
5.2 Filter and Z-Transform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
5.3 Intuitive Filter Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
5.4 Relation with Fourier Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 262
5.5 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
6 Filter Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
6.1 Filter Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
6.2 FIR Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
6.3 IIR Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395
6.4 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 494
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 500
7 Implementation Matters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 503
7.1 Number Representations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 504
7.2 Filter Implementation Components . . . . . . . . . . . . . . . . . . . . . . . 516
7.3 FIR Filter Implementations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 553
7.4 IIR Filter Implementations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 558
7.5 Frequency Domain Filter Realization . . . . . . . . . . . . . . . . . . . . . . 563
7.6 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 593
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 594
8 Filters with MATLAB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 595
8.1 Fundamental Filter Design Methods . . . . . . . . . . . . . . . . . . . . . . . 595
8.2 Advanced Filter Design Methods . . . . . . . . . . . . . . . . . . . . . . . . . 603
8.3 MATLAB Code Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 637
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 643

Appendix A: MATLAB Fundamentals . . . . . . . . . . . . . . . . . . . . . . . . . . . 645

Appendix B: MATLAB Symbolic Math Toolbox . . . . . . . . . . . . . . . . . . . 661
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 673
About the Author

Dr. Keonwook Kim received the B.S. degree in Electronics Engineering from
Dongguk University, Seoul, Korea in 1995 and the M.S. and Ph.D. degrees in
Electrical and Computer Engineering from the University of Florida, Gainesville,
United States in 1997 and 2001, respectively. He is presently Professor in the
Division of Electronics and Electrical Engineering at the Dongguk University. Prior
to joining Dongguk University, he worked as Assistant Professor in the Department
of Electrical and Computer Engineering at the Florida State University from 2001
to 2003. His primary research interest is acoustic localization via using the
multi-aural architecture in order to mimic the aural system of animals which include
human.

xiii
Chapter 1
Preliminary Digital Filter Design

The general filter passes various matters to separate out unwanted things. In elec-
trical engineering, the filter is known as the device for minimizing or suppressing
the noise frequencies to obtain the better signals. The digital filter handles the input
and output in discrete time domain as well as the quantized magnitude form. This
chapter introduces the digital filter from scratch. The filter equation is derived from
the general engineering concept. We assume that the input signal to the digital filter
presents the real-time property which represents the past, present, and future tense
in the signal. The conceptual derivation performed in this chapter actually operates
filter function in the real field based on the trial and error approach. In the following
chapters, the abstract design is followed by the further delicate analyses to meet the
accurate filter performance.

1.1 Discrete Time Signal

The continuous time is t as real number and discrete time is n that is the integer
number. The () handles the continuous arguments and [] deals with discrete argu-
ments. Any functions can be continuous and discrete form with arguments. The
given signal is x[n] as shown in Fig. 1.1.
The discrete signal with integer argument has no tense as past, present, and
future unless we specify the current time. That is the sequence of signal which
shows the relative position of numbers. The higher index number shows later, and
the lower index appears earlier. The real-time signal with n index provides the
present time position n which is the variable integer number. Therefore, the n + 1,
n + 2, and etc. are the future of the signal and n−1, n−2 … are the past of them.
Based on the time schedule, we only have the data up to the n index and n + 1 data
will be gathered next. The magnitude of the data is real number and any numbers
are possible for received value. The frequency components of the signal are not
numerically analyzed yet, but we presume that the signal with rapid fluctuation
© Springer Nature Singapore Pte Ltd. 2021 1
K. Kim, Conceptual Digital Signal Processing with MATLAB,
Signals and Communication Technology,
https://doi.org/10.1007/978-981-15-2584-1_1
2 1 Preliminary Digital Filter Design

1 2 ..
.
0

Fig. 1.1 a Example of discrete time signal with absolute time index number. b The corresponding
discrete time signal with real-time representation with n. The clock shows the current time n

contains the additional high frequency components. In contrast, the signal with
gentle movement includes the dominant low frequency components. The source of
the signal can be anything such as dice toss and heart rate. Conventionally the
discrete time signal is obtained from the continuous information by regular sam-
pling. We assume that the sampling distance for time is even; hence, the actual time
difference between adjacent indexes are always identical. After every fixed time, the
number for the discrete index n will be updated to the next integer. In the digital
signal processing, the x[n] has two meanings those are whole signal sequence
(variable n) and current signal value (given n). The discrete signal can be defined as
the signal sequence as shown below.

p 
x½n ¼ cos n ð1:1Þ
2

If the sequence is used to build another discrete time signal, the input sequence is
employed as real-time signal as shown below.

1 1 1
y½n ¼ x½n þ x½n  1 þ x½n  2 ð1:2Þ
3 3 3

The y[n] function uses the current x value as x[n] and two previous values as x[n
−1] and x[n−2].
1.2 Design the Digital Filters 3

1.2 Design the Digital Filters

The real-time discrete time signal is given, and you are asked to design the filter to
reduce the noise frequencies and emphasize signal information. I believe that this is
the initial point of digital signal processing. Let’s design the filter.
The filter output y[n] is the operated results from the filter input x[n]s, and the
signal is obtained up to the n index value, as shown above Fig. 1.2. Any idea? Note
that the signal is continuously received by the system in every constant interval;
therefore, the system is willing to have the significant amount of data in short time.
It is impossible to process the whole obtained data for filtering; hence, the limited
length of data is considered to provide the filter output. The simple idea is to take
average the recent inputs for low pass filtering (smoothing). The recent N samples
are denoted as below.

x½n; x½n  1; x½n  2; . . .; x½n  N þ 1 ð1:3Þ

The current filter out y[n] is computed as below.

1
y½n ¼ fx½n þ x½n  1 þ x½n  2 þ    þ x½n  N þ 1g ð1:4Þ
N

The filter output with input data is illustrated in Fig. 1.3. The recent three data
samples are considered to create average value for output y[n]. In Fig. 1.3, the
n values (n is the variable and not the current index here) are demonstrated from

Present

Future
Gathered data

n-5 n-4 n-3 n-2 n-1 n n+1 n+2 n+3 n+4 n+5

Fig. 1.2 One example of discrete time signal with real-time representation
4 1 Preliminary Digital Filter Design

Fig. 1.3 Digital average filter with recent three data

seven to ten and the output values are noticeably smoothed comparing to the
income signal. This is the primitive low pass filter (LPF).
Example 1.1
Write the equation for digital average filter with recent three data.
Solution

1
y½n ¼ fx½n þ x½n  1 þ x½n  2g
3
∎
The independent parameter for the LPF is operation (averaging) and length
(three). As shown in Eq. (1.4), the averaging operation can be seen as the weighted
sum for the recent incoming data. Instead of using the averaging computation, the
different weight values can be applied on the data set. First of all, let’s change the
length of the averaging in the LPF as Fig. 1.4.
Example 1.2
Write the equation for digital average filter with recent seven data.
Solution

1
y½n ¼ fx½n þ x½n  1 þ    þ x½n  5 þ x½n  6g
7
∎
1.2 Design the Digital Filters 5

Fig. 1.4 Digital average filter with recent seven data

For the increased length (seven) of averaging operation, the filter output y
[n] delivers further smoothed outcome than the shorter version (length three).
Therefore, depending on the filtering length, smoothness of the output can be
decided proportionally. In other word, the longer length averaging passes the
narrower range frequencies from zero frequency, and the shorter length averaging
passes the wider range frequencies that demonstrates the high fluctuation in filter
output. The length of the filter is important parameter to determine the frequency
range of the output.
Example 1.3
Write the equation for y[7] digital average filter in Fig. 1.5.
Solution

1
y½7 ¼ fx½7 þ x½6 þ    þ x½2 þ x½1g
7

∎
The other parameter to be considered is weight values in the filter. The averaging
operation multiplies the constant 1/N to each data values for limited N length and
accumulates for filter output. The averaging range, latest N data, can be seen as the
window with weight for the given time sequence, and the window slides to the next
for new output. Figure 1.5 illustrates the explained procedures in terms of window
with seven window length. We understand that the filter length controls the output
6 1 Preliminary Digital Filter Design

Fig. 1.5 Digital average filter with recent seven data

frequency range. The shape or weight of the window significantly contributes to

manipulate the frequency component of the output as well. Let’s do another
experiment.
Example 1.4
Write the equation for high pass filter in Fig. 1.6.
Solution

1
y½n ¼ fx½n þ x½n  1  x½n  2g
3

∎
Unlike averaging filter, the filter output shows the higher variation in magnitude
than the original signal; therefore, the high frequency components are exaggerated via
the filtering process. The designed filter is high pass filter (HPF) with three window
length. The values in the window decide the characteristics of the filter which prefers
to pass the high frequency components; hence, we can think that the window shape
plays an important role. Comparing to the LPF, the HPF also uses the same window
length but the filter output is completely reverse in action. Applying the longer HPF
window derives same variation as LPF? The longer HPF length generates concen-
trated filter output to the high frequencies than the three HPF length situation as
Fig. 1.7.
1.2 Design the Digital Filters 7

Fig. 1.6 High pass filter with recent three data

Fig. 1.7 High pass filter with recent seven data

8 1 Preliminary Digital Filter Design

Example 1.5
Write the equation for high pass filter in Fig. 1.7.
Solution

1
y½n ¼ fx½n þ x½n  1 þ    þ x½n  5  x½n  6g
7

∎
The window length controls the frequency focus of the filter output in inversely
proportional manner. The longer window creates the further concentrated output for
the specific filter type which is determined by the window shape. The relationship
between the output frequencies and window length is derived from the above
simple experiments. How we can decide the filter type from window shape? Let’s
draw the two previously used filters in Fig. 1.8.
Can you figure out the filter type by observing the filter shape? Yes, the filter
output follows the window shape; therefore, smoothness and roughness of the
window provide the filter type that specify the designated frequency you want to
pass. Other than LPF and HPF, let’s perform another example. The window shape
of this filter is certain periodic signal which shows the 6-sample period and
19-sample length. The filter output is similar to the window shape as illustrated in
Fig. 1.9. The longer length expects to emphasize the window shape on the output
based on the intuition of previous lessons.

Fig. 1.8 a Low pass filter weight. b High pass filter weight
1.2 Design the Digital Filters 9

Fig. 1.9 The filter to pass the 6-sample period signal

Example 1.6
Write the equation for digital filter in Fig. 1.9.
Solution

X
18  
2pk
y½n ¼ x½n  k cos
k¼0
6
∎
Up to now, we have designed primitive filters for LPF, HPF, etc. This is the
fundamental of the digital filter theory. The idea and computation initiated the
filtering system and signal processing in digital domain. The design method based
on the trial and error does not lend the solid foundation for building the intended
filter. The shown intuitive methods should be formularized in engineering area for
further analysis and application. This book not only provides the mathematical
representations of the algorithms but also illustrates the physical meanings of the
system extensively in the coming sections and chapters.

1.3 Filter Architecture

The illustrated filter system provides approximate methodological principle of the

filtering. This section converts the pictured procedures into the mathematical for-
mula. Let’s consider the LPF as below.
10 1 Preliminary Digital Filter Design

1 1 1 1
y½n ¼ x½n þ x½n  1 þ x½n  2 þ    þ x½n  N þ 1 ð1:5Þ
N N N N

The recent N samples of data sequence is averaged for the low pass filtering in
terms of weighting the 1/N on each sample. Since the weight values in the window
can be various for intended filtering purpose, the window is separated and slid
sequentially in every interval as Fig. 1.10. The signal sequence x[n] starts from the
zero n value and the window will be overlapped at the zero index. As the time goes
by, overlap range will be increased and completed after N−1 sample shift; hence,
we expect to have proper averaged output from the operations.
The window function is specified as below.

1
q½0 ¼ q½1 ¼ q½2 ¼    ¼ q½ðN  1Þ ¼
N

Therefore, the first appropriate output by complete overlap is defined as;

y½N  1 ¼ q½ðN  1Þx½0 þ q½ðN  2Þx½1 þ q½ðN  3Þx½2 þ   

þ q½1x½N  2 þ q½0x½N  1

The second and third proper outputs are shown below with sliding input
sequence.

y½N ¼ q½ðN  1Þx½1 þ q½ðN  2Þx½2 þ q½ðN  3Þx½3 þ   

þ q½1x½N  1 þ q½0x½N

x[n]

... n
0 1 2 ...

q[n]
Sliding
1/N
... n
-(N-1) 0

Sum Sum
Sliding
... ... n
... ... n
Output
Output
y[n] y[n]
... n
... n

Fig. 1.10 The low pass filtering operation. The x[n] is the input sequence and the q[n] is the filter
weight window
1.3 Filter Architecture 11

y½N þ 1 ¼ q½ðN  1Þx½2 þ q½ðN  2Þx½3 þ q½ðN  3Þx½4 þ   

þ q½1x½N þ q½0x½N þ 1

Generalizing the above equation by summation as below.

X
N1
y½n ¼ x½n  k q½k ð1:6Þ
k¼0

The n is the current time in the equation and the integer value n starts from zero
to infinite. Based on the recent N input sequence, the filter output is computed by
weight q[] function. This is the nice equation to formulate the intuitive concept of
the primitive filtering. Since the input data sequence initiates from the zero-time
index to positive number, the weight window is also flipped over the vertical axis in
order to start from the zero as below.

q½0 ! h½0; q½1 ! h½1; q½2 ! h½2; . . .; q½ðN  1Þ ! h½N  1

Due to the relocation, the tense of the weight window is changed completely as
shown in Fig. 1.11. The original window function q[] represents the q[0] as present
moment and the its tails on the left as past time. Note that the q[0] is always
multiplied with the present x[] value in the filtering computations shown at
Eq. (1.6). The further to the left indicates the earlier times in weight window. In the
flipped window function h[], the further to the right is past and h[0] is present time.
Therefore, the equation with h[] is below.

y½n ¼h½ðN  1Þx½n  ðN  1Þ þ h½ðN  2Þx½n  ðN  2Þ þ   

þ h½3x½n  2 þ h½1x½n  1 þ h½0x½n

... q[n] h[n] ...

0 0

Past Future Future Past

Flipping
Present Present
Fig. 1.11 Flipping the filter weight window for h[n]
12 1 Preliminary Digital Filter Design

Fig. 1.12 Digital filter block diagram

Generalizing the above equation by summation as below.

X
N 1
y½n ¼ x½n  k h½k ð1:7Þ
k¼0

The only difference with non-flipped window filter is that there is no minus sign
in the window function. This is the digital filter equation also known as convolution
sum. The equation is the time domain filter with given input signal x[] and pro-
duces the filtered output y[]. For the specific purpose, the shape and length of h[]
have to be determined in analytical manner to improve the signal and reduce the
noise component from incoming discrete data sequence on Fig. 1.12.
The diagram for the filter is given as above. The real-time input sequence x[] is
provided to the filter system represented by h[]. The corresponding filter output is y
[].

1.4 Digital Filter Definitions and Requirements

By definition, the filter is a process to eliminate the unwanted frequencies, which

are classified as noise, from an input signal. The designed filters were explained in
terms of emphasizing the certain frequencies; however, a range of frequencies are
suppressed to intensify the information in fact. The important thing in the filter
realization is that ‘do not add any information by filtering.’ You can increase and
decrease certain frequency magnitudes but cannot create new frequencies in the
filtering process. This is the reason that the filter is explained by elimination of
frequencies.
One example of digital signal processing is shown in Fig. 1.13. The input signal
is the cosine wave with certain period and the processing output limits the input
magnitude to one. The input and output frequency distributions are illustrated as
well. The input frequency consists of single dominant component and the output
frequency presents at least three components. The additional prominent frequency
components are pointed by the circles in the Fig. 1.13. The given digital signal
processing is disqualified from the filter requirements according to the frequency
component non-creation rule. However, the reverse situation is valid as the digital
1.4 Digital Filter Definitions and Requirements 13

Fig. 1.13 An example of digital signal processing

filter. The output sequence placed at the filter input provides the output which only
includes the single frequency component. The multiple frequency components are
removed to obtain the signal frequency; therefore, the filter dose not insert any
frequencies to the output.
Example 1.7
Write the equation to add the zero frequency component over the signa x[n].
Solution

y½n ¼ x½n þ constant

Since the constant values indicate the zero frequency, the output y[n] contains
the additional zero frequency component.
∎
Now we know that the designed filter by h[] can compute the output by
Eq. (1.7). When you have the unknown filter, how we can find the h[] and what is
the relationship with y[] from h[]? What signal figures out the h[]? Do not look at
the equation. Observe that we are dealing with discrete time signal. The primitive
and fundamental element of the discrete signal is the function that has single value
in the certain time index. Let’s see the example at Fig. 1.14.
14 1 Preliminary Digital Filter Design

q[n]
2
1
2
n
01
-1

=
q0[n] q1[n] q2[n]
2
1
+ + 2
n n n
012 012 01
-1

Fig. 1.14 An example of function decomposition

The function q[] is the linear combination (simple addition) of q0[], q1[] and
q2[]. The individual q1[] and q2[] can be represented by the shift location and
magnitude multiplication of q0[] as shown below.

Shift right by 1 Multiply by 2

q0 ½n ! ! q1 ½n

Shift right by 2 Multiply by 1

q0 ½n ! ! q2 ½n

Therefore, the manipulations and combinations of q0[] can provide the q[] and
any arbitrary functions. Based on this idea, we can think about the filter output by
analytical combinations. If the filter output of q[] is the linear combination of the
filter outputs from q0[], q1[] and q2[], then the q[] output can be decomposed into
the simpler form. Furthermore, the individual filter output of q1[] and q2[] can be
described by the shift location and magnitude multiplication of q0[] output just as
relationships of q0[], q1[] and q2[], then the filter output will be expressed by
linear combination of representative output like q0[] in this example. The gener-
ating output from individual inputs is illustrated in Fig. 1.15.
Figure 1.15 shows that the filter output can be derived from the linear combi-
nation and time shift of the primitive output. This kind of the system is called as
linear and time invariant system. The linear system denotes the filter outcome by
addition over scaled version of individual element outputs. The time invariant
system preserves the filter output shape for the time shifted input with identical time
relocation. With linear and time invariance condition, the system can be charac-
terized by very simple format that is the output of the single value in single time.
1.4 Digital Filter Definitions and Requirements 15

Fig. 1.15 The linear q0[n]

combination of shifted output
y0[n]
from decomposed input for
arbitrary input response 1 1
n Response
n
012 01234

+ +
q1[n] y1[n]
2 2

n Response
n
012 01234

q2[n]
+ +
y2[n]

2
n Response n
01 01234
-1 -1

q[n] y[n] 3
2 2
1 1 1
2
n Response
01234
n
01
-1 -1

Any other outputs can be produced by the scaled and shifted combinations of the
characterizing output. This is very important property since the characterizing
output represents the complete feature of the system described as h[] in the
equation. How we can find the h[] from the unknown linear and time invariant
system? Let’s perform Fig. 1.10 example in reverse manner as below.
As shown in Fig. 1.16, the impulse signal (one magnitude at time zero), which is
the characterizing input, sequentially scans the h[] of the system and provides the h
[] of the filter. The discovered h[] tells us the property and performance of the filter
in detail. Therefore, the user can find the system feature of the unknown and linear
16 1 Preliminary Digital Filter Design

x[n]

n
0 1 2 ...

h[n]

0 1234 n

x[n] x[n] x[n] x[n] x[n]

1 1 1 1 1

n n n n n
0 1 2 ... 0 1 2 ... 0 1 2 ... 0 1 2 ... 0 1 2 ...

h[-n]

-4 -3-2-1 0 1 2 3 4 n -4 -3-2-1 0 1 2 3 4 n -4 -3-2-1 0 1 2 3 4 n -4 -3-2-1 0 1 2 3 4 n -4 -3-2-1 0 1 2 3 4 n

y[n] y[n] y[n] y[n] y[n]

01234 n 01234 n 01234 n 01234 n 01234 n

Sliding

Fig. 1.16 Place the impulse signal to derive the h[n]

& time invariant filter by impulse signal input. Now, we understand that the h[] can
be designed and found for the specific filtering purpose. However, we have to
explore further to study the relationship between the h[] and frequency components
in coming chapters.

1.5 What We Need for Further (Optional)

If you are going to build the digital filter by using the trial and error method, the
chapter one is enough information for you. The information above shows about
window shape and filter computation intuitively. The low pass filter is realized by
the constant weight window for filter equation and the length is determined by your
trials until you obtain the satisfactory outcome. Other types of the filters can be
implemented by the derived window and numerous executions. The fundamental
concept is delivered but certain analytical approaches are missed in above.
For the engineering aspects, the systematical methodology is required for
approximate closed form solutions in filter design. The frequency is fundamental
elements of the system input/output and the signal is the limited or unlimited
combination of frequency components in general. The definition of the frequency is
introduced in the first part of this book. The digital signal processing manages the
signal in discrete time domain; however, the frequency is well understood in the
continuous time domain. The frequency for the digital signal is explained with the
sampling theory that describes the signal transformation between the continuous
and discrete time domain. The digital signal can be decomposed into the simple
1.5 What We Need for Further (Optional) 17

elements with linear and time shifting combination; therefore, the fundamental and
essential signal elements are described and defined as well.
Typically, the given signals are delivered in time domain and do not represent
the frequency information directly. To understand the signal and analyze the per-
formance, the designer needs to see the spectral distribution in the signal. The
Fourier analysis provides the mathematical tool to transform the information
between the time and frequency. The Fourier analysis computes the magnitude and
phase (or delay) of the frequencies from the periodic and non-periodic signal.
However, the Fourier analysis does not generate all-round solutions to all transform
matters. The users need to note that there are numerous limitations and conditions to
apply the analysis over the signals. The physical meaning and overall constraints of
Fourier analysis are described in this book.
The time domain filter is already introduced in this chapter. The filter equation is
primitive but essential fundamental of the signal frequency managing. Upon the
understanding of the frequency definition and Fourier analysis, the filter equation is
revisited to comprehend further based on the mathematical study. The frequency
response of the filter from the window shape and length is derived to meet the
intended filter requirements. Also, the finite window is extended to the infinite
length by recursive filter architecture and its frequency response is explored as well.
By using the frequency modulation, the low pass filter can be located in any
frequency position for designated filter specification. The simple filters based on the
frequency modulation are provided for practical filter design.
The Fourier analysis can be extended to the Z-transform to understand the signal
and system with additional perspective. The Z-transform converts the subject
between the time domain and the Z domain known as complex number plane. The
rational polynomial of complex number Z represents limited and unlimited length
signal and system with constant coefficients. The solutions to the polynomial
provide the various information such as filter type, stability, and etc. The beauty of
the Z-transform is that the transform presents powerful mathematical tool to handle
and visualize the system in simple manner. Also, the Z-transform can be adopted to
design the any type filters.
The digital filter is implemented by the digital processor (or logic) with analog to
digital converter (ADC) and digital to analog converter (DAC). Since the capability
of the processor and ADC/DAC are limited, the filter algorithm and signal repre-
sentation should be optimized in terms of computation and dimension. Numerous
concerns should be exercised to realize the digital filter in real-time processing. In
the software perspective, the algorithm can be executed quickly via using the
corresponding fast algorithms. In the other hardware viewpoint, the numbers can be
handled in the fixed- or floating-point representation that can change the execution
accuracy and speed. Therefore, the algorithm structure and format significantly
affect the filter performance in various ways.
This book extensively uses the MATLAB to understand the digital signal pro-
cessing theory throughout the chapters. In the last chapter, the comprehensive
Exploring the Variety of Random
Documents with Different Content
nations shall not be barricaded by her or by others. Meanwhile, the
absolute independence of Egypt has become a necessity. Her
connection with Turkey is an unmixed and unmitigated evil. She
wants all her hands; and yet she was compelled to send a large
contingent to the last Russo-Turkish campaign. When the Turks are
freely allowed wholesale 'repudiation,' the Nile Valley must pay her
usurious creditors the uttermost farthing of a public debt individually
contracted. She wants all her money, and she should, in common
justice, be freed from her heavy tribute, and from the heavier
benevolences and douceurs which perpetually find their way to the
Seraglio, or rather into ministerial pockets. And Egypt is at present,
despite the rose-water reports of officials, who take a personal pride
in writing, not the truth but what is wished to be the truth, far from
comfortable. Abyssinia has placed her between the horns of a
dilemma. She must either grant or not grant a port to the barbarous
and bloodthirsty Nestorians called Christians. In the former case, the
only imports will be arms and ammunition, especially the latter, for
Johannes, the Emperor, has thousands of breech-loaders, but no
cartridges. In the latter case she will be in a state of chronic war
with her turbulent neighbour, who is ever threatening her inland
frontier.
"And Egypt has lately offended the moral sense of Europe by a
peculiarly retrograde Mohammedan measure—the systematic revival
of the import slave-trade. England has a manner of convention with
her for suppressing it: but the provisions of 1877 should be made
more stringent. She has now reached that point of civilization when
she can afford to proclaim a total abolition of compulsory labour, the
full and immediate emancipation of the 'chattel.' Slaves and
eunuchs, the latter denounced by Islamism, are mere articles of
luxury for pashas and beys. I will not deny that when the infamous
revival of trade in human beings was brought to their notice, her
Ministry addressed a circular letter to the Mudirs or provincial
Governors, and appointed a director for its suppression. But this,
again, was the Eastern trick of poudre aux yeux. The director went
up the Nile, and the slaves came down the Red Sea. Then the
director, having apparently done enough for a rose-water report,
retired to his winter quarters at Cairo, and the slaves returned to the
Nile. Meanwhile the Ministry, whilst permitting this shameful traffic,
has systematically neglected the gold and silver placers discovered
on the Midian coast, and evidently extending far southwards; in fact,
the old Ophir and Havilah. In Turkish Arabia (the vilayet province of
Yemen, near Sana'a), a new digging has been discovered, and, with
true Oriental exaggeration, has been proclaimed 'one of the richest
in the world.' But in the hands of the Turkish Government even a
diamond-mine, a Golconda, would be a losing affair; it can be
worked with profit only by European heads and hands. Meanwhile
Egypt must recover her prestige by abolishing slavery and by
exploiting her mineral wealth.
"To conclude. Poets are sometimes prophets; and we have a
specimen in the forecast of Camoens, which dates from the year of
grace 1572—
'Those fierce projectiles, of our days the work,
Murderous engines, dire artilleries,
Against Byzantine walls, where dwells the Turk,
Should long ago have belcht their batteries.
Oh, hurl it back, in forest caves to lurk,
Where Caspian crests and steppes of Scythia freeze,
That Turkish ogre-progeny multiplied
By potent Europe's policy and pride.'

"What also wrote Torquato Tasso, only a few years after Camoens?
'For if the Christian Princes ever strive
To win fair Greece out of the tyrant's hands,
And those usurping Ismaelites deprive
Of woeful Thrace, which now captived stands;
You must from realms and sea the Turks forth drive,
As Godfrey chased them from Judah's lands,' etc.
Amen, and so be it!
"R. F. B.
"Trieste."

[1] Written in 1876.

[2] Lord Beaconsfield.
[3] This was written in 1876.
[4] "This was written at the end of 1876. It would be impossible to-day (1878) not
to sympathize with and admire Austria and her brave army struggling single-
handed and manfully in the great Bosnian and Herzegovinian difficulty, but when it
is over her reward will be great. It is a large step in the right direction; but we,
who want a great Austrian Empire, wish she had had all the nineteen million Slavs,
not a part."
[5] "This was written January, 1876."
[6] "I fear that the Future now threatens to be the Present (1893)."—I. B.
[7] This was written in 1876.
[8] His grandson was Chancellier at Trieste in 1888.—I. B.
[9] "See 'Lettres, Journal, et Documents,' vol. ii. pp. 298-300. He rates the mob at
five thousand, and writes dramatically. The cushions of a divan do not form an
espèce de tombeau, where a woman can be ensevelie vivante. M. de Lesseps says
that he had the details from the chief actors of the drama, but I prefer M.
Sabatier's account."
[10] "Our diamond weights are as follows:—
16 parts = 1 (diamond) grain = 4/5 grain, troy.
4 diamond grains = 1 carat = 3 1/5 (3.174 grains, troy).
"The Indian weights are:—
1 Dhan = 15.32 grains, troy, in round numbers half a grain.
4 Dhary = 1 Rati = 1 2/3 grains, troy.
8 Rati = 1 Masha = 18 grains, troy.
12 Mashas = 1 Tola = 180 grains troy.
"The 'ounces' in the text probably represents 'tolas,' certainly not troy ounces of
24 grains."
[11] "Mr. Maclean kindly drew my attention to the Treaty with the Nizam
(November 12th, 1766), which cedes to the E. I. Company 'the five Circars or
Provinces of Ellour (Ellore, north of Masulipatam), Rajahmondra Siccacole (or
Chicacole on the coast), and Moortizanuggur or Gunton.' The four first named
were added to the French dominions by De Bussy. 'These Circars,' we read,
'include territory extending along the coast from the mouths of the Kistna
(Krishna) northward to near Ganjour, and stretching some distance inland.' Article
No. 11 of the same Treaty runs thus: 'The Hon'ble E. I. Company, in consideration
of their diamond mines, with the villages appertaining thereto, having been always
dependent on H. H. the Nizam's Government, do hereby agree that the same shall
remain in possession now also.'"
[12] "All this was written two years before the late Afghan War began."

APPENDIX F.

LETTERS ON THE JEDDAH MASSACRE, AND CHOLERA—HIS

WARNING TO THE GOVERNMENT, WHICH CALLED DOWN A
REPRIMAND ON HIM.

"To the Secretary of the Royal Geographical Society, London.

"Sir,—I have the honour to inform you that on the 1st of December,
1856, I addressed to you a letter which I hope has been duly
received. On the 2nd instant, in company with Lieutenant Speke, I
left Bombay Harbour, on board the H.E.I. Company's ship of war
Elphinstone (Captain Frushard, I.N., commanding), en route to East
Africa. I have little to report that may be interesting to geographers;
but perhaps some account of political affairs in the Red Sea may be
deemed worthy to be transmitted by you to the Court of Directors or
to the Foreign Office.
"As regards the Expedition, copies of directions and a memorandum
on instruments and observations for our guidance have come to
hand. For observations, Lieutenant Speke and I must depend upon
our own exertions, neither serjeants nor native students being
procurable at the Bombay Observatory. The case of instruments and
the mountain barometer have not been forwarded, but may still find
us at Zanzibar. Meanwhile I have obtained from the Commanding
Engineer, Bombay, one six-inch sextant, one five and a half ditto,
two prismatic compasses, five thermometers (of which two are B.P.),
a patent log, taper, protractors, stands, etc.; also two pocket
chronometers from the Observatory, duly rated; and Dr. Buist,
secretary Bombay Geographical Society, has obliged me with a
mountain barometer and various instructions about points of
interest. Lieutenant Speke has been recommended by the local
Government to the Government of India for duty in East Africa, and
the services of Dr. Steinhaüser, who is most desirous to join us, have
been applied for from the Medical Board, Bombay. I have strong
hopes that both these officers will be allowed to accompany me, and
that the Royal Geographical Society will use their efforts to that
effect.
"By the subjoined detailed account of preliminary expenses at
Bombay, it will be seen that I have expended £70 out of £250 for
which I was permitted to draw.
"Although, as before mentioned, the survey of Eastern Intertropical
Africa has for the moment been deferred, the necessity still exists.
Even in the latest editions of Horsburgh, the mass of matter relative
to Zanzibar is borrowed from the observations of Captain Bissel, who
navigated the coast in H.M.'s ships Leopard and Orestes, about A.D.
1799. Little is known of the great current which, setting periodically
from and to the Red Sea and the Persian Gulf, sweeps round the
Eastern Horn of Africa. The reefs are still formidable to navigators;
and before these seas can be safely traversed by steamers from the
Cape, as is now proposed, considerable additions must be made to
Captain Owen's survey in A.D. 1823-24. Finally, operations on the
coast will form the best introduction to the geographical treasures of
the interior.
"The H.E.I. Company's surveying brig Tigris will shortly be out of
dock, where she has been undergoing a thorough repair, and if fitted
up with a round-house on the quarter-deck would answer the
purpose well. She might be equipped in a couple of months, and
despatched to her ground before the south-west monsoon sets in, or
be usefully employed in observing at Zanzibar instead of lying idle in
Bombay Harbour. On former surveys of the Arabian and African
Coasts, a small tender of from thirty to forty tons has always been
granted, as otherwise operations are much crippled in boisterous
weather and exposed on inhospitable shores. Should no other vessel
be available, one of the smallest of the new pilot schooners now
unemployed at Bombay might be directed to wait upon the Tigris.
Lieutenant H. G. Fraser, I.N., has volunteered for duty upon the
African coast, and I have the honour to transmit his letter. Nothing
more would be required were some junior officer of the Indian navy
stationed at Zanzibar for the purpose of registering tidal, barometric,
and thermometric observations, in order that something of the
meteorology of this unknown region may be accurately investigated.
"When passing through Aden I was informed that the blockade of
the Somali Coast had been raised without compensation for the
losses sustained on my last journey. This step appears, politically
speaking, a mistake. In the case of the Mary Ann brig, plundered
near Berberah in A.D. 1825, due compensation was demanded and
obtained. Even in India, an officer travelling through the states not
under British rule can, if he be plundered, require an equivalent for
his property. This is, indeed, our chief protection—semi-barbarians
and savages part with money less willingly than with life. If it be
determined for social reasons at Aden that the blockade should
cease and mutton become cheap, a certain percentage could be paid
upon the exports of Berberah till such time as our losses, which,
including those of Government, amount to £1380, are made good.
"From Harar news has reached Aden that the Amir Abubakr, dying
during the last year of chronic consumption, has been succeeded by
a cousin, one Abd el Rahman, a bigoted Moslem, and a violent hater
of the Gallas. His success in feud and foray, however, has not
prevented the wild tribes from hemming him in, and unless fortune
interferes, the city must fall into their hands. The rumour prevalent
at Cairo, namely, that Harar had been besieged and taken by Mr.
Bell, now serving under 'Theodorus, Emperor of Ethiopia' (the chief
Cássái), appears premature. At Aden I met in exile Sharmarkay bin
Ali Salih, formerly Governor of Zayla. He has been ejected in favour
of a Dankali chief by the Ottoman authorities of Yemen—a
circumstance the more to be regretted as he has ever been a firm
friend to our interests.
"The present defenceless state of Berberah still invites our presence.
The eastern coast of the Red Sea is almost entirely under the Porte.
On the western shore, Cosseir is Egyptian; Masáwwah, Sawakin, and
Zayla, Turkish; and Berberah, the best port of all, unoccupied. I have
frequently advocated the establishment of a British agency at this
place, and venture to do so at once. This step would tend to
increase trade, to obviate accidents in case of shipwreck, and
materially assist in civilizing the Somal of the interior. The
Government of Bombay has doubtless preserved copies of my
reports, plans, and estimates concerning the proposed agency, and I
would request the Royal Geographical Society to inquire into a
project peculiarly fitted to promote their views of exploration in the
Eastern Horn of Africa. Finally, this move would checkmate any
ambitious projects in the Red Sea. The Suez Canal may be said to
have commenced. It appears impossible that the work should pay in
a commercial sense. Politically it may, if at least its object be, as
announced by the Count d'Escayrac de Lauture, at the Société de
Géographie, to 'throw open the road of India to the Mediterranean
coasting trade, to democratize commerce and navigation.' The first
effect of the highway would be, as that learned traveller justly
remarks, to open a passage through Egypt to the speronari and
feluccas of the Levant, the light infantry of a more regular force.
"The next step should be to provide ourselves with a more efficient
naval force at Aden, the head-quarters of the Red Sea squadron. I
may briefly quote, as a proof of the necessity for protection, the
number of British protégés in the neighbouring ports, and the
present value of the Jeddah trade.
"Mocha now contains about twenty-five English subjects, the
principal merchants in the place. At Masáwwah, besides a few
French and Americans, there are from sixteen to twenty British
protégés, who trade with the interior, especially for mules required at
the Mauritius and our other colonies. Hodaydah has from fifty to
sixty, and Jeddah, besides its dozen resident merchants, annually
witnesses the transit of some hundreds of British-protected subjects,
who flock to the Haj for commerce and for devotion.
"The chief emporium of the Red Sea trade for centuries past has
been Jeddah, the port of Meccah. The custom-house reports of 1856
were kindly furnished to me by Captain Frushard, I.N. (now
commanding the H.E.I.C.'s sloop of war Elphinstone), an old and
experienced officer, lately employed in blockading Berberah, and
who made himself instrumental in quelling certain recent attempts
upon Turkish supremacy in Western Arabia. According to these
documents, thirty-five ships of English build (square-rigged) arrived
at and left Jeddah between the end of September and April, from
and for various places in the East, China, Batavia, Singapore,
Calcutta, Bombay, the Malabar coast, the Persian Gulf, and Eastern
Africa. Nearly all carried our colours, and were protected, or
supposed to be protected, by a British register: only five had on
board a European captain or sailing master, the rest being
commanded and officered by Arabs and Indians. Their cargoes from
India and the Eastern regions are rice, sugar, piece goods, planking,
pepper, and pilgrims; from Persia, dates, tobacco, and raw silk; and
from the Mozambique, ivory, gold dust, and similar costly articles.
These imports in 1856 are valued at £160,000. The exports for the
year, consisting of a little coffee and spice for purchase of imports,
amounts, per returns, to £120,000. In addition to these square-
rigged ships, the number of country vessels, open boats, bungalows,
and others, from the Persian Gulf and the Indian coasts, amounts to
900, importing £550,000, and exporting about £400,000. I may
remark, that to all these sums at least one-third should be added, as
speculation abounds, and books are kept by triple entry in the Holy
Land.
"The next port in importance to Jeddah is Hodaydah, where vessels
touch on their way northward, land piece and other goods, and call
on the return passage to fill with coffee. As the head-quarters of the
Yemen Pashalik, it has reduced Mocha, formerly the great coffee
mart, to insignificance, and the vicinity of Aden, a free port, has
drawn off much of the stream of trade from both these ancient
emporia. On the African coast of the Red Sea, Sawakin, opposite
Jeddah, is a mere slave mart, and Masáwwah, opposite Hodaydah,
still trades in pearls, gold dust, ivory, and mules.
"But if the value of the Red Sea traffic calls, in the present posture of
events, for increased means of protection, the slave-trade has equal
claims to our attention. At Aden energetic efforts have been made to
suppress it. It is, however, still carried on by her country boats from
Sawakin, Tajarrah, Zayla, and the Somali coast; a single cargo
sometimes consisting of two hundred head gathered from the
interior, and exported to Jeddah and the small ports lying north and
south of it. The trade is, I believe, principally in the hands of Arab
merchants at Jeddah and Hodaydah, and resident foreigners,
principally Indian Moslems, who claim our protection in case of
disturbances, and consequently carry on a thriving business. Our
present squadron in the Red Sea consisting of only two sailing-
vessels, the country boats in the African ports have only to wait till
they see the ship pass up or down, and then, knowing the passage
—a matter of a day—to be clear, to lodge the slaves at their
destination. During the past year, this trade was much injured by the
revolt of the Arabs against the Turks, and the constant presence of
the Elphinstone, whose reported object was to seize all vessels
carrying slaves. The effect was principally moral. Although the
instructions for the guidance of the Commander enjoined him to
carry out the wishes of the Home and Indian Governments for the
suppression of slavery, yet there being no published treaty between
the Imperial Government and the Porte sanctioning to us the right of
search in Turkish bottoms, his interference would not have been
supported by the Ottoman local authorities. It may be well to state,
that after a Firman had been published in the Hejaz and Yemen
abolishing the trade, the Turkish Governments of Jeddah and
Hodaydah declared that the English Commander might do as he
pleased, but that they declined making any written request for his
assistance. For its present increased duties, for the suppression of
the slave-trade, for the protection of British subjects, and for the
watching over Turkish and English interests in the Red Sea, the Aden
Squadron is no longer sufficient. During the last two years it has
numbered two sailing-vessels—the Elphinstone, a sloop of war,
carrying twelve 32-pounders and two 12-pounders; and the Mahi, a
schooner armed with one pivot gun, 32-pounder, and two 12-
pounders. Nor would it be benefited by even a considerable increase
of sailing-vessels. It is well known that, as the prevailing winds
inside the sea are favourable for proceeding upwards from
September to April, so on the return, during those months, they are
strongly adverse. A fast ship, like the Elphinstone, requires thirty
days on the downward voyage to do the work of four. Outside the
sea, during those months, the current sets inward from the Indian
Ocean, and a ship, in event of very light winds falling, has been
detained a whole week in sight of Aden. From April to September, on
the contrary, the winds set down the Red Sea frequently with
violence, the current inside the sea also turns towards the Indian
Ocean, and outside the south-west monsoon is blowing. Finally,
sailing-ships draw too much water. In the last year the Elphinstone
kept the Arabs away from Jeddah till the meanness of the Sherif Abd
el Muttalib had caused his downfall. But her great depth (about from
14'6 to 15 feet) prevented her approaching the shore at Hodaydah
near enough to have injured the insurgents, who, unaware of the
fact, delayed their attack upon the town till famine and a consequent
pestilence dispersed them. With little increase of present
expenditure, the Red Sea might be effectually commanded. Two
screw-steamers, small enough to enter every harbour, and to work
steadily amongst the banks on either shore, and yet large enough to
be made useful in conveying English political officers of rank and
native princes, when necessary, would amply suffice. A vessel of the
class of H.M.'s gunboat Flying Fish, drawing at most nine feet of
water, and carrying four 32-pounders of 25 cwt. each, as broadside,
and two 32-pounders of 25 cwt. each, as pivot guns, would probably
be that selected. The crews would consist of fewer men than those
at present required, and means would easily be devised for
increasing the accommodation of officers and men, and for securing
their health and comfort during cruises that might last two months in
a hot and dangerous climate.
"By means of two such steamers we shall, I believe, be prepared for
any contingencies which may arise in the Red Sea; and if to this
squadron be added an allowance for interpreters and a slave
approver in each harbour—in fact, a few of the precautions practised
by the West African Squadron—the slave-trade in the Red Sea will
soon have received its death-blow, and Eastern Africa its
regeneration at our hands.
"I have, etc., etc.,
"R. F. B,
"Commanding East African Expedition.
"H.E.I.C. Sloop of War Elphinstone,
"15th December, 1856."

The Official "Judicious" (?) Reply.

No. 961 of 1857.

From H. L. Anderson, Esquire, Secretary to Government, Bombay, to

Captain R. F. Burton, 18th Regiment Bombay N.I.
"Dated the 23rd July, 1857.
"Sir,—With reference to your letter, dated the 15th December, 1856,
to the address of the Secretary of the Royal Geographical Society of
London, communicating your views on affairs in the Red Sea, and
commenting on the political measures of the Government of India, I
am directed by the Right Honourable the Governor in Council to
state, your want of discretion, and due respect for the authorities to
whom you are subordinate, has been regarded with displeasure by
Government.
"I have the honour to be, Sir,
"Your most obedient Servant,
"(Signed) H. L. Anderson,
"Secretary to Government.
"Bombay Castle, 23rd July, 1857."

[Richard received by same post as above letter the account of the

Massacre at Jeddah.]

(Extracts from the Telegraph Courier, Overland Summary,

Bombay, August 4, 1858.)

"On the 30th of June, a massacre of nearly all the Christians took
place at Jeddah on the Red Sea. Amongst the victims were Mr. Page,
the British Consul, and the French Consul and his lady. Altogether
the Arabs succeeded in slaughtering about twenty-five.
"H.M. steamship Cyclops was there at the time, and the captain
landed with a boat's crew, and attempted to bring off some of the
survivors, but he was compelled to retreat, not without having killed
a number of the Arabs. The next day, however, he succeeded in
rescuing the few remaining Christians, and conveyed them to Suez.
"Amongst those who were fortunate enough to escape was the
daughter of the French Consul; and this she succeeded in doing
through the fidelity of a native, after she had killed two men with
her own hands, and been severely wounded in the encounter.
Telegraphic despatches were transmitted to England and France, and
the Cyclops is waiting orders at Suez. As it was apprehended that
the news from Jeddah might excite the Arab population of Suez to
the commission of similar outrages, H.B.M.'s Vice-Consul at that
place applied to the Pasha of Egypt for assistance, which was
immediately afforded by the landing of five hundred Turkish soldiers,
under the orders of the Pasha of Suez."

"Unyanyembe, Central Africa, 24th June, 1858.

"Sir,—I have the honour to acknowledge the receipt of your official
letter, No. 961 of 1857, conveying to me the displeasure of the
Government in consequence of my having communicated certain
views on political affairs in the Red Sea to the Royal Geographical
Society of Great Britain.
"The paper in question was as is directly stated, and it was sent for
transmission to the Board of Directors, or the Foreign Office, not for
publication. I beg to express my regret that it should have contained
any passage offensive to the authorities to whom I am subordinate;
and to assure the Right Honourable the Governor in Council that
nothing was further from my intentions than to displease a
Government to whose kind consideration I have been, and am still,
so much indebted.
"In conclusion, I have the honour to remind you that I have received
no reply to my official letter, sent from Zanzibar, urging our claims
upon the Somal for the plunder of our property.
"I have the honour to be, Sir,
"Your most obedient Servant,
"Richard F. Burton,
"Commanding East African Expedition.
"To the Secretary to Government, Bombay."
 No. 2845 of 1857. Political Department.

From H. L. Anderson, Esq., Secretary to Government of Bombay, to

Captain R. F. Burton, Commanding E. A. Expedition, Zanzibar.
"Dated 13th June, 1857.
"Sir,—I am directed by the Right Honourable the Governor in Council
to acknowledge the receipt of your letter dated the 26th April last,
soliciting compensation on behalf of yourself and other members of
the late Somalee Expedition, for losses sustained by you and them.
"2. In reply, I am desired to inform you, that under the opinion
copied in the margin [here reproduced below], expressed by the late
Governor-General of India, the Right Honourable the Governor in
Council cannot accede to the application now preferred.
"I have, etc.,
"(Signed) H. L. Anderson,
"Secretary to Government."
[From margin.] "Having regard to the conduct of the Expedition, his
Lordship cannot think that the officers who composed it have any
just claims on the Government for their personal losses."

END OF FIRST CORRESPONDENCE.

[Here begins the Speke and Rigby cabal.]

Second Correspondence.

1.

"India Office, E.C., November 8th, 1859.

"Sir,—I am directed by the Secretary of State for India in Council to
forward for your information, copy of a letter addressed by Captain
Rigby, her Majesty's Consul and Agent at Zanzibar, to the
Government of Bombay, respecting the non-payment of certain
persons hired by you to accompany the Expedition under your
command into Equatorial Africa, and to request that you will furnish
me with any observations which you may have to make upon the
statements contained in that letter.
"Sir Charles Wood especially desires to be informed why you took no
steps to bring the services of the men who accompanied you, and
your obligations to them, to the notice of the Bombay Government.
"I am, Sir,
"Your obedient servant,
"(Signed) T. Cosmo Melvill.
"Captain R. Burton."

2.

No. 70 of 1859. Political Department.

From Captain C. P. Rigby, her Majesty's Consul and British Agent,

Zanzibar, to H. L. Anderson, Esquire, Secretary to Government,
Bombay.
"Zanzibar, July 15th, 1859.
"Sir,—I have the honour to report, for the information of the Right
Honourable the Governor in Council, the following circumstances
connected with the late East African Expedition under the command
of Captain Burton.
"2. Upon the return of Captain Burton to Zanzibar in March last, from
the interior of Africa, he stated that, from the funds supplied him by
the Royal Geographical Society for the expenses of the Expedition,
he had only a sufficient sum left to defray the passage of himself
and Captain Speke to England, and in consequence the persons who
accompanied the Expedition from here, viz. the Kafila Bashi, the
Belooch sepoys, and the porters, received nothing whatever from
him on their return.
"3. On quitting Zanzibar for the interior of Africa, the Expedition was
accompanied by a party of Belooch soldiers, consisting of a Jemadar
and twelve armed men. I understand they were promised a monthly
salary of five dollars each; they remained with the Expedition for
twenty months, and as they received nothing from Captain Burton
beyond a few dollars each before starting, his Highness the Sultan
has generously distributed amongst them the sum of two thousand
three hundred (2300) dollars.
"4. The head clerk of the Custom House here, a Banian, by name
Ramjee, procured ten men, who accompanied the Expedition as
porters. They were promised five dollars each per mensem, and
received pay for six months, viz. thirty dollars each before starting
for the interior. They were absent for twenty months, during three of
which the Banian Ramjee states that they did not accompany the
Expedition. He now claims eleven months' pay for each of these
men, as they have not been paid anything beyond the advance
before starting.
"5. The head clerk also states that, after the Expedition left Zanzibar,
he sent two men to Captain Burton with supplies, one of whom was
absent with the Expedition seventeen months, and received nothing
whatever; the other, he states, was absent fifteen months, and
received six months' pay, the pay for the remaining nine months
being still due to him. Thus his claim amounts to the following sum:
—
 Dollars.
Ten men for eleven at five dollars per man per
550
months, month
One man for seventeen " " 85
One man for nine " " 45
Total 680
"6. These men were slaves, belonging to 'deewans,' or petty chiefs,
on the opposite mainland. They travel far into the interior to collect
and carry down ivory to the coast, and are absent frequently for the
space of two or three years. When hired out, the pay they receive is
equally divided between the slave and the master. Captain Speke
informs me, that when these men were hired, it was agreed that
one-half of their hire should be paid to the men, and the other half
to Ramjee on account of their owners. When Ramjee asked Captain
Burton for their pay, on his return here, he declined to give him
anything, saying that they had received thirty dollars each on
starting, and that he could have bought them for a less sum.
"7. The Kafila Bashi, or chief Arab, who accompanied the Expedition,
by name Said bin Salem, was twenty-two months with Captain
Burton. He states that on the first journey to Pangany and
Usumbara, he received fifty (50) dollars from Captain Burton; and
that before starting on the last Expedition, to discover the Great
Lake, the late Lieutenant-Colonel Hamerton presented him with five
hundred dollars on behalf of Government for the maintenance of his
family during his absence. He states that he did not stipulate for any
monthly pay, as Colonel Hamerton told him that if he escorted the
gentlemen to the Great Lake in the interior, and brought them in
safety back to Zanzibar, he would be handsomely rewarded, and
both Captain Speke and Mr. Apothecary Frost inform me that Colonel
Hamerton frequently promised Said bin Salem that he should receive
a thousand dollars and a gold watch if the Expedition were
successful.
"8. As it appeared to me that Colonel Hamerton had received no
authority from Government to defray any part of the expenses of
this Expedition, and probably made these promises, thinking that if
the exploration of the unknown interior were successful a great
national object would be attained, and that the chief man who
conducted the Expedition would be liberally rewarded, and as
Captain Burton had been furnished with funds to defray the
expenses, I told him that I did not feel authorized to make any
payment without the previous sanction of Government, and Said bin
Salem has therefore received nothing whatever since his return.
"9. Said bin Salem also states, that on the return of the Expedition
from Lake Tanganyika, seventy (70) natives of the country were
engaged as porters, and accompanied the Expedition for three
months; and that on arriving at a place called 'Kootoo,' a few days'
journey from the sea-coast, Captain Burton wished them to diverge
from the correct route to the coast opposite Zanzibar, to accompany
him south to Keelwa; but they refused to do so, saying that none of
their people ever dared to venture to Keelwa, where the chief slave-
trade on the east coast is carried on. No doubt their fears were well
grounded. These men received nothing in payment for their three
months' journey, and, as no white man had ever penetrated into
their country previously, I fear that any future traveller will meet
with much inconvenience in consequence of these poor people not
having been paid.
"10. As I considered that my duty connected with the late Expedition
was limited to affording it all the aid and support in my power, I
have felt very reluctant to interfere with anything connected with the
non-payment of these men; but Said bin Salem and Ramjee having
appealed to me, and Captain Speke, since his departure from
Zanzibar, having written me two private letters, pointing out so
forcibly the claims of these men, the hardships they endured, and
the fidelity and perseverance they showed, conducting them safely
through unexplored countries, and stating also that the agreements
with them were entered into at the British Consulate, and that they
considered they were serving the British Government, that I deem it
my duty to bring their claims to the notice of Government; for I feel
that if these men remain unpaid, after all they have endured in the
service of British officers, our name for good faith in these countries
will suffer, and that any future traveller wishing to further explore
the interesting countries of the interior will find no persons willing to
accompany them from Zanzibar, or the opposite mainland.
"11. As there was no British agent at Zanzibar for thirteen months
after the death of Colonel Hamerton, the Expedition was entirely
dependent on Luddah Damha, the Custom-master here, for money
and supplies. He advanced considerable sums of money without any
security, forwarding all requisite supplies, and, Captain Speke says,
afforded the Expedition every assistance in the most handsome
manner. Should Government, therefore, be pleased to present him
with a shawl, or some small mark of satisfaction, I am confident he
is fully deserving of it, and it would gratify a very worthy man to find
that his assistance to the Expedition is acknowledged.
"I have, etc.,
"(Signed) C. P. Rigby, Captain,
"H.M.'s Consul and British Agent, Zanzibar."

3.

"East India United Service Club, St. James's Square,

"November 11th, 1859.
"Sir,—I have the honour to acknowledge the receipt of your official
letter, dated the 8th of November, 1859, forwarding for my
information copy of a letter, addressed by Captain Rigby, her
Majesty's Consul and Agent at Zanzibar, to the Government of
Bombay, respecting the non-payment of certain persons hired by me
to accompany the Expedition under my command into Equatorial
Africa, and apprising me that Sir C. Wood especially desires to be
informed why I took no steps to bring the services of the men who
accompanied me, and my obligations to them, to the notice of the
Bombay Government.
"In reply to Sir Charles Wood, I have the honour to state that, as the
men alluded to rendered me no service, and as I felt in no way
obliged to them, I would not report favourably of them. The Kafilah
Bashi, the Jemadar, and the Beloch were servants of H.H. Sayyid
Majid, in his pay and under his command. They were not hired by
me, but by the late Lieut.-Colonel Hamerton, H.M.'s Consul and
H.E.I.C.'s Agent at Zanzibar, and they marched under the Arab flag.
On return to Zanzibar, I reported them as undeserving of reward to
Lieut.-Colonel Hamerton's successor, Colonel Rigby, and after return
to England, when my accounts were sent in to the Royal
Geographical Society, I appended a memorandum, that as those
persons had deserved no reward, no reward had been applied for.
"Before proceeding to reply to Captain Rigby's letter, paragraph by
paragraph, I would briefly premise with the following remarks:—
"Being ordered to report myself to Lieut.-Colonel Hamerton, and
having been placed under his direction, I admitted his friendly
interference, and allowed him to apply to H.H. the Sultan for a guide
and an escort. Lieut.-Colonel Hamerton offered to defray, from public
funds, which he understood to be at his disposal, certain expenses
of the Expedition, and he promised, as reward to the guide and
escort, sums of money, to which, had I been unfettered, I should
have objected as exorbitant. But in all cases the promises made by
the late Consul were purely conditional, depending entirely upon the
satisfactory conduct of those employed. These facts are wholly
omitted in Captain Rigby's reports.
"2. Captain Rigby appears to mean that the Kafila Bashi, the Beloch
sepoys, and the porters received nothing whatever on my return to
Zanzibar, in March last, from the interior of Africa because the funds
supplied to me by the Royal Geographical Society for the
expenditure of the Expedition, had been exhausted, besides the sum
of one thousand pounds (£1000) granted by the Foreign Office. I
had expended from my own private resources nearly fourteen
hundred pounds (£1400), and I was ready to expend more had the
expenditure been called for. But, though prepared on these
occasions to reward liberally for good service, I cannot see the
necessity, or rather I see the unadvisability, of offering a premium to
notorious misconduct. This was fully explained by me to Captain
Rigby on my return to Zanzibar.
"3. Captain Rigby 'understands' that the party of Beloch sepoys,
consisting of a Jemadar and twelve armed men, were promised a
monthly salary of five dollars each. This was not the case. Lieut.-
Colonel Hamerton advanced to the Jemadar twenty-five, and to each
sepoy twenty dollars for an outfit; he agreed that I should provide
them with daily rations, and he promised them an ample reward
from the public funds in case of good behaviour. These men
deserved nothing; I ignore their 'fidelity' and 'perseverance,' and I
assert that if I passed safely through an unexplored country, it was
in no wise by their efforts. On hearing of Lieut.-Colonel Hamerton's
death, they mutinied in a body. At the Tanganyika Lake they refused
to escort me during the period of navigation, a month of danger and
difficulty. When Captain Speke proposed to explore the Nyanza Lake,
they would not march without a present of a hundred dollars' worth
of cloth. On every possible occasion they clamoured for bakshish,
which, under pain of endangering the success of the Expedition,
could not always be withheld. They were often warned by me that
they were forfeiting all hopes of a future reward, and, indeed, they
ended by thinking so themselves. They returned to Zanzibar with a
number of slaves, purchased by them with money procured from the
Expedition. I would not present either guide or escort to the Consul;
but I did not think it my duty to oppose a large reward, said to be
2300 dollars, given to them by H.H. the Sultan, and I reported his
liberality and other acts of kindness to the Bombay Government on
my arrival at Aden. This fact will, I trust, exonerate me from any
charge of wishing to suppress my obligations.
"4. The Banyan Ramjee, head clerk of the Custom House, did not, as
is stated by Captain Rigby, procure me ten (10) men who
accompanied the Expedition as porters; nor were these men, as is
asserted (in par. 6), 'slaves belonging to deewans or petty chiefs on
the opposite mainland.' It is a notorious fact that these men were
private slaves, belonging to the Banyan Ramjee, who hired them to
me direct, and received from me as their pay, for six months, thirty
dollars each; a sum for which, as I told him, he might have bought
them in the bazaar. At the end of six months I was obliged to
dismiss these slaves, who, as is usually the case with the slaves of
Indian subjects at Zanzibar, were mutinous in the extreme. At the
same time, I supplied them with cloth, to enable them to rejoin their
patron. On my return from the Tanganyika Lake, they requested
leave to accompany me back to Zanzibar, which I permitted, with the
express warning that they were not to consider themselves re-
engaged. The Banyan, their proprietor, had, in fact, sent them on a
trading trip into the interior under my escort, and I found them the
most troublesome of the party. When Ramjee applied for additional
pay, after my return to Zanzibar, I told him that I had engaged them
for six months; that I had dismissed them at the end of six months,
as was left optional to me; and that he had already received an
unusual sum for their services. This conversation appears in a
distorted form and improperly represented in the concluding
sentence of Captain Rigby's 6th paragraph.
"5 and 6. With respect to the two men sent on with supplies after
the Expedition had left Zanzibar, they were not paid, on account of
the prodigious disappearance of the goods entrusted to their charge,
as I am prepared to prove from the original journals in my
possession. They were dismissed with their comrades, and never
afterwards, to the best of my remembrance, did a day's work.
"7 and 8. The Kafilah Bashi received from me for the first journey to
Usumbara fifty (50) dollars. Before my departure in the second
Expedition he was presented by Lieut.-Colonel Hamerton with five
hundred (500) dollars, almost double what he had expected. He was
also promised, in case of good conduct, a gold watch, and an ample
reward, which, however, was to be left to the discretion of his
employers. I could not recommend him through Captain Rigby to the
Government for remuneration. His only object seemed to be that of
wasting our resources and of collecting slaves in return for the heavy
presents made to the native chiefs by the Expedition, and the
consequence of his carelessness or dishonesty was, that the
expenditure on the whole march, until we had learnt sufficient to
supervise him, was inordinate. When the Kafilah Bashi at last
refused to accompany Captain Speke to the Nyanza Lake, he was
warned that he also was forfeiting all claim to future reward, and
when I mentioned this circumstance to Captain Rigby at Zanzibar, he
then agreed with me that the 500 dollars originally advanced were
sufficient.
"9. With regard to the statement of Said bin Salim concerning the
non-payment of the seventy-three porters, I have to remark that it
was mainly owing to his own fault. The men did not refuse to
accompany me because I wished to diverge from the 'correct route,'
nor was I so unreasonable as to expect them to venture into the
jaws of the slave-trade. Several caravans that had accompanied us
on the down-march, as well as the porters attached to the
Expedition, were persuaded by the slaves of Ramjee (because
Zanzibar was a nearer way to their homes) not to make Kilwa. The
pretext of the porters was simply that they would be obliged to
march back for three days. An extra remuneration was offered to
them; they refused it, and left in a body. Shortly before their
departure Captain Speke proposed to pay them for their services,
but being convinced that they might be prevented from desertion, I
did not judge it advisable by paying them, to do what would be
virtually dismissing them. After they had proceeded a few miles, Said
bin Salim was sent to recall them, on conditions which they would
have accepted; he delayed, lost time, and ended by declaring that
he could not travel without his dinner. Another party was instantly
sent; they also loitered on the way, and thus the porters reached the
coast and dispersed. Before their departure I rewarded the
Kirangozi, or chief man of the caravan, who had behaved well in
exhorting his followers to remain with us. I was delayed in a most
unhealthy region for the arrival of some down porters, who
consented to carry our goods to the coast; and to prove to them
that money was not my object, I paid the newly engaged gang as if
they had marched the whole way. Their willingness to accompany
me is the best proof that I had not lost the confidence of the people.
Finally, on arrival at the coast, I inquired concerning those porters
who had deserted us, and was informed by the Diwan and headman
of the village that they had returned to their homes in the interior,
after a stay of a few days on the seaboard. This was a regrettable
occurrence, but such events are common on the slave-path in
Eastern Africa, and the established custom of the Arabs and other
merchants, whom I had consulted upon the subject before leaving
the interior, is not to encourage desertion by paying part of the hire,
or by settling for porterage before arriving at the coasts. Of the
seven gangs of porters engaged on this journey, only one, an
unusually small proportion, left me without being fully satisfied.
"10. That Said bin Salim, and Ramjee, the Banyan, should have
appealed to Captain Rigby, according to the fashion of Orientals,
after my departure from Zanzibar, for claims which they should have
advanced when I refused to admit them, I am not astonished. But I
must express my extreme surprise that Captain Speke should have
written two private letters, forcibly pointing out the claims of these
men to Captain Rigby, without having communicated the
circumstance in any way to me, the chief of the Expedition. I have
been in continued correspondence with that officer since my
departure from Zanzibar, and until this moment I have been
impressed with the conviction that Captain Speke's opinion as to the
claims of the guide and escort above alluded to was identical with
my own.
"11. With respect to the last paragraph of Captain Rigby's letter,
proposing that a shawl or some small mark of satisfaction should be
presented by Government to Ladha Damha, the Custom-master at
Zanzibar, for his assistance to the Expedition, I distinctly deny the
gratuitous assertions that I was entirely dependent on him for
money and supplies; that he advanced considerable sums of money
without any security; that he forwarded all requisite supplies, or, as
Captain Speke affirms, that he afforded the Expedition every
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

textbookfull.com