Download the full version and explore a variety of ebooks

or textbooks at https://ebookultra.com

Channel Coding Theory Algorithms and Applications

1st Edition David Declercq (Editor)

_____ Tap the link below to start your download _____

https://ebookultra.com/download/channel-coding-theory-
algorithms-and-applications-1st-edition-david-declercq-
editor/

Find ebooks or textbooks at ebookultra.com today!

 We believe these products will be a great fit for you. Click
the link to download now, or visit ebookultra.com
to discover even more!

Sparse Modeling Theory Algorithms and Applications 1st

Edition Irina Rish

https://ebookultra.com/download/sparse-modeling-theory-algorithms-and-
applications-1st-edition-irina-rish/

Elements of quasigroup theory with applications in coding

and cryptology 1st Edition Shcherbacov

https://ebookultra.com/download/elements-of-quasigroup-theory-with-
applications-in-coding-and-cryptology-1st-edition-shcherbacov/

Fundamentals of wavelets Theory algorithms and

applications 2ed Edition Goswami J.

https://ebookultra.com/download/fundamentals-of-wavelets-theory-
algorithms-and-applications-2ed-edition-goswami-j/

Speech Coding Algorithms Foundation and Evolution of

Standardized Coders 1st Edition Wai C. Chu

https://ebookultra.com/download/speech-coding-algorithms-foundation-
and-evolution-of-standardized-coders-1st-edition-wai-c-chu/
Coding theory a first course Ling

https://ebookultra.com/download/coding-theory-a-first-course-ling/

Image fusion algorithms and applications Tania Stathaki

https://ebookultra.com/download/image-fusion-algorithms-and-
applications-tania-stathaki/

Channel Adaptive Technologies and Cross Layer Designs for

Wireless Systems with Multiple Antennas Theory and
Applications 1st Edition Vincent K.N. Lau
https://ebookultra.com/download/channel-adaptive-technologies-and-
cross-layer-designs-for-wireless-systems-with-multiple-antennas-
theory-and-applications-1st-edition-vincent-k-n-lau/

Introduction to coding theory 2nd Edition Bierbrauer

https://ebookultra.com/download/introduction-to-coding-theory-2nd-
edition-bierbrauer/

Bayesian Phylogenetics Methods Algorithms and Applications

1st Edition Ming-Hui Chen

https://ebookultra.com/download/bayesian-phylogenetics-methods-
algorithms-and-applications-1st-edition-ming-hui-chen/
Channel Coding Theory Algorithms and Applications 1st
Edition David Declercq (Editor) Digital Instant Download
Author(s): David Declercq (editor), Marc Fossorier (editor), Ezio Biglieri
(editor)
ISBN(s): 9780081013304, 0081013302
Edition: 1
File Details: PDF, 19.53 MB
Year: 2014
Language: english
 Channel Coding
Theory, Algorithms, and
Applications

Academic Press Library

in Mobile and Wireless
Communications
 Channel Coding
Theory, Algorithms, and
Applications

Edited by

DAVID DECLERQ, MARC FOSSORIER

AND EZIO BIGLIERI

Academic Press Library in Mobile

and Wireless Communications

AMSTERDAM • BOSTON • HEIDELBERG • LONDON

NEW YORK • OXFORD • PARIS • SAN DIEGO
SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO
Academic Press is an imprint of Elsevier
Academic Press is an imprint of Elsevier
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK
225 Wyman Street, Waltham, MA 02451, USA

First edition 2014

Copyright © 2014 Elsevier Ltd. All rights reserved

No part of this publication may be reproduced, stored in a retrieval system or transmitted in

any form or by any means electronic, mechanical, photocopying, recording or otherwise with-
out the prior written permission of the publisher.

Permissions may be sought directly from Elsevier’s Science & Technology Rights Department
in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; email: permis-
[email protected]. Alternatively you can submit your request online by visiting the Elsevier
web site at http://elsevier.com/locate/permissions, and selecting Obtaining permission to use
Elsevier material.

Notice
No responsibility is assumed by the publisher for any injury and/or damage to persons or
property as a matter of products liability, negligence or otherwise, or from any use or opera-
tion of any methods, products, instructions or ideas contained in the material herein. Because
of rapid advances in the medical sciences, in particular, independent verification of diagnoses
and drug dosages should be made.

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress
ISBN–13: 978-0-12-396499-1

For information on all Academic Press publications

visit our web site at store.elsevier.com

Printed and bound in the US

14 15 16 17 18 10 9 8 7 6 5 4 3 2 1
Preface

Channel coding has long been recognized as an important feature for the transmis-
sion or the storage of digital information, to combat the unstructured noise incurred
by small to nano-electronics. Data are always used in coded form for their transmis-
sion through wireless or wired channels, or for their storage on magnetic or opti-
cal recording devices, to ensure a desired level of reliability of the communication
systems.
While its origin dates back to more than half a century, it was not until the 1990s
and the introduction of the turbo principle in information theory that theoretical lim-
its could be approached with practical designs. This discovery, based on the concept
of iterative decoding with feedback of newly processed information, revolutionized
the area of forward error correction (FEC) and nowadays every modern communica-
tions or storage system has been designed with the consideration of an iterative FEC
scheme. Consequently, wireless communications have been directly impacted by
iterative decoding methods. The turbo principle also motivated the designs of other
iteratively decodable codes and in particular contributed to the resurrection of low
density parity check (LDPC) codes. The iterative information processing gives rise to
so strong practical algorithms that the concept has also been applied for more general
communication systems and receivers, incorporating feedback loops between the
FEC and the diverse signal processing blocks. We can cite as most popular examples
receivers implementing turbo-equalization, turbo-detection, turbo-synchronization,
etc. The efficient design of those iterative receivers takes its root in the deep under-
standing of the theory of turbo and LDPC codes.
In this book, we review the concepts of channel coding relevant to wireless commu-
nications in conjunction with the designs that heavily rely on iterative decoding meth-
ods. Although an impressive leap has been achieved in the performance of FEC for
wireless communications due to iterative methods, issues about their design and imple-
mentation remain. In fact, the initial gains achieved were so large compared to previous
FEC schemes that at first, almost every iterative decoding scheme that you could think
of seemed good. These days, refinements about these designs in terms of structure,
complexity, latency, etc. are still of importance, both theoretically and practically.
The authors who contributed to this chapter are leading experts in the area they
cover, with a deep knowledge of both the latest theory and recent practical realiza-
tions of the topic, as well as of the remaining issues. In each chapter, an emphasis is
made on the presentation of the concepts and the most efficient and useful techniques,
with insistence on heavily referencing the corresponding literature. Consequently,
this book is intended to address a large audience from practical engineers to research-
ers to graduate students.
This book is formed of 13 chapters, which can be divided into four conceptual
parts: Part I (Chapters 1 to 4) describes the main iteratively decodable codes; Part II
(Chapters 5 to 7) covers tools to design these codes in practical implementations; Part
III (Chapters 8 to 12) presents the combination of these codes with other techniques
xv
xvi Preface

relevant to wireless communications. Finally, Part IV (Chapter 13) addresses the

VLSI realization of these schemes.
In Chapter 1, the original turbo codes are first reviewed, with emphasis on the
concepts associated with the initial design, put in a historical perspective, and an
analysis of the performance based on constituents of a turbo code. Finally, the indus-
trial impacts of the turbo codes are presented.
Chapter 2 covers design, analysis, construction, and performance of the turbo-
like codes. This chapter also describes the iterative decoding algorithms for these
codes, with many simulation results provided.
Chapter 3 introduces LDPC codes and their constructions. It first describes the
important parameters for these codes based on an asymptotic analysis. Then con-
structions of practical importance for implementation are presented, with a final
description of LDPC codes already in standards.
Chapter 4 first presents an overall survey of LDPC decoders, and then provides a
more detailed insight into some of the most widely used decoders, with emphasis on
their implementation.
In Chapter 5, one of the most important tools to design iteratively decodable
codes, namely, the extrinsic information transfer (EXIT) method, is presented from a
practical point of view. It is shown how to apply it to design several classes of itera-
tively decodable codes.
Chapter 6 presents a study of the causes of decoding failures on various channels
under different iterative decoding algorithms.
In Chapter 7, practical considerations of FEC such as puncturing with rate com-
patibility and unequal error protection are presented for iteratively decodable codes.
Chapter 8 presents the concept of rateless coding and the main categories of rate-
less codes that have been proposed for various types of channel. Applications for
which rateless coding can be used are then discussed.
Chapter 9 surveys distributed channel coding techniques for cooperative commu-
nications, with emphasis on decode-and-forward relaying. Several classes of codes
are discussed.
Chapter 10 introduces the important aspects of space–time block codes (STBCs)
related to their encoding, decoding, rate diversity, and diversity-multiplexing trade-
offs. Many construction techniques for STBCs are presented.
Chapter 11 considers coded modulation, which jointly combines coding and mod-
ulation to increase bandwidth efficiency. Both original designs and recent designs
based on iteratively decodable codes are presented.
Chapter 12 provides an overview of joint source-channel coding and decoding,
as well as of the related problem of joint protocol-channel decoding techniques. The
impact of these techniques in actual systems is finally discussed.
Chapter 13 gives a systematic review of the main achievements in the implemen-
tation of iterative channel decoders. Solutions required to achieve specific design
objectives, such as low complexity, high throughput, and low energy, are discussed.
In conclusion, the editors would like to thanks all authors for their time and efforts
in order to meet the high expectations of this project.
Contributors
Humberto Beltrão
School of Engineering and Science, Jacobs University gGmbH, Campus Ring 1,
D-28725 Bremen, Germany
Sergio Benedetto
Dipartimento di Elettronica e Telecomunicazioni (DET), Politecnico di Torino
C.so Duca degli Abruzzi n. 24, 10129 Torino, Italy
Emmanuel Boutillon
Université de Bretagne Sud, Lab-STICC, Centre de Recherche, BP 92216,
56321 Lorient Cedex, France
Shashi Kiran Chilappagari
Marvell Semiconductor Inc., Santa Clara, CA 95054, USA
Dariush Divsalar
Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak Grove Drive,
MS 238–420, Pasadena, CA, USA
Catherine Douillard
Telecom Bretagne, UMR6285, Lab-STICC, F29200 Brest, France
Pierre Duhamel
LSS – SUPELEC 3, 3, rue Jolio Curie, 91192, Gif-Sur-Yvette, France
Mark Flanagan
School of Electrical, Electronic and Communications Engineering,
University College Dublin, Belfield, Dublin 4, Ireland
Alexandre Graell i Amat
Department of Signals and Systems, Chalmers University of Technology,
Gothenburg, Sweden
Werner Henkel
School of Engineering and Science, Jacobs University gGmbH, Campus Ring 1,
D-28725 Bremen, Germany
Motohiko Isaka
Department of Informatics, School of Science and Technology,
Kwansei Gakuin University, Japan
Michel Jezequel
Telecom Bretagne, UMR6285, Lab-STICC, F29200 Brest, France
Michel Kieffer
LSS – SUPELEC 3, 3, rue Jolio Curie, 91192, Gif-Sur-Yvette, France

xvii
xviii Contributors

Ingmar Land
Institute for Telecommunications Research, University of South Australia,
Mawson Lakes, SA 5095, Australia
Guido Masera
Politecnico di Torino, Department of Electronics and Telecommunications,
corso Duca degli Abruzzi 24, 10129 Torino, Italy
Guido Montorsi
Dipartimento di Elettronica e Telecomunicazioni (DET), Politecnico di Torino
C.so Duca degli Abruzzi n. 24, 10129 Torino, Italy
Dung Viet Nguyen
Department of Electrical and Computer Engineering, University of Arizona,
Tucson, AZ 85721, USA
Enrico Paolini
Department of Electrical, Electronic and Information Engineering “G. Marconi”,
University of Bologna, Via Venezia 52, Cesena, FC 47521, Italy
C. Poulliat
IRIT Lab, INP/ENSEEIHT-Toulouse, 2 rue Charles Camichel, B.P. 7122,
31071 Toulouse Cedex 7, France
Aditya Ramamoorthy
Iowa State University, 3222 Coover Hall, Department of Electrical and
Computer Engineering, Ames, IA 50010, USA
Valentin Savin
CEA-LETI, MINATEC Campus, 17 rue des Martyrs, 38054 Grenoble, France
B. Sundar Rajan
Department of ECE, IISc, Bangalore 560012, India
Ragnar Thobaben
School of Electrical Engineering, Royal Institute of Technology (KTH),
Stockholm, Sweden
Bane Vasić
Department of Electrical and Computer Engineering, University of Arizona,
Tucson, AZ 85721, USA
 CHAPTER

Turbo Codes: From First

Principles to Recent
Standards
Catherine Douillard and Michel Jezequel
1
Telecom Bretagne, UMR6285, Lab-STICC, F29200 Brest, France

CHAPTER OUTLINE
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2 History of turbo codes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2.1 The origins of turbo codes . . . . . . . . . . . . . . . . . . . . . . . . 3
2.2 Concatenation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.3 Negative feedback in the decoder and recursive systematic
convolutional codes . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.4 Extrinsic information and iterative decoding . . . . . . . . . . . . . . 4
2.5 Parallel concatenation . . . . . . . . . . . . . . . . . . . . . . . . . . 6
3 Fundamentals of turbo coding . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.1 Recursive systematic convolutional (RSC) component codes . . . . . . 8
3.2 Block coding with turbo codes . . . . . . . . . . . . . . . . . . . . . . 13
3.3 The permutation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.3.1 Regular permutation . . . . . . . . . . . . . . . . . . . . . . . . 19
3.3.2 Irregular permutations . . . . . . . . . . . . . . . . . . . . . . . 23
4 Fundamentals of turbo decoding . . . . . . . . . . . . . . . . . . . . . . . . 26
4.1 The turbo principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4.2 Soft-input soft-output decoding . . . . . . . . . . . . . . . . . . . . . 30
4.2.1 Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
4.2.2 The MAP algorithm . . . . . . . . . . . . . . . . . . . . . . . . 32
4.2.3 The MAP algorithm in the logarithmic domain:
Log-MAP and Max-Log-MAP . . . . . . . . . . . . . . . . . . . . 36
5 Industrial impacts of turbo codes . . . . . . . . . . . . . . . . . . . . . . . . 38
5.1 The very first implementations of turbo codecs . . . . . . . . . . . . . 38
5.1.1 The CAS 5093 circuit . . . . . . . . . . . . . . . . . . . . . . . 39
5.1.2 The Turbo4 circuit . . . . . . . . . . . . . . . . . . . . . . . . . 40
5.2 Early applications of turbo codes . . . . . . . . . . . . . . . . . . . . 41
5.3 Turbo codes in standards . . . . . . . . . . . . . . . . . . . . . . . . . 43

Channel Coding: Theory, Algorithms, and Applications. http://dx.doi.org/10.1016/B978-0-12-396499-1.00001-7

© 2014 Elsevier Ltd. All rights reserved.
1
2 CHAPTER 1 Turbo Codes: From First Principles to Recent Standards

5.3.1 Mobile communication systems . . . . . . . . . . . . . . . . . . 43

5.3.2 Digital video broadcasting (DVB) standards . . . . . . . . . . . . 44
5.3.3 Other standards . . . . . . . . . . . . . . . . . . . . . . . . . . 45
5.3.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

1 Introduction
This chapter is a general introduction to the original turbo codes, proposed and
patented by Claude Berrou in 1991 [1–3] and known as convolutional turbo codes or
parallel concatenated convolutional codes. Turbo codes are an outcome of the research
activity of Telecom Bretagne1 (formerly École Nationale des Télécommunications de
Bretagne), a French graduate engineering school in the field of information technolo-
gies. The Electronics Department of Telecom Bretagne has been involved in research
in the field of algorithm-silicon interaction for more than 25 years. Its activity mainly
consists in jointly devising new algorithms and innovative hardware architectures,
digital and analog, for digital communications.
The chapter describes the main concepts of coding theory introduced with the
invention of turbo codes, provides the fundamental guidelines for the design of turbo
codes with good performance, gives the basics of turbo decoding, and briefly reviews
the industrial impacts of this new generation of error-correcting codes. Most of the
sections are introduced from a historical point of view. The chapter is organized as
follows. Section 2 describes the experimentations, observations, and reflections which
led to turbo codes and the ensuing concepts of iterative decoding, extrinsic informa-
tion, and parallel concatenation. Section 3 focuses on the different constituents of the
turbo encoder and analyzes their effects on the code performance. Section 4 provides
the basics of the turbo principle and soft-input soft-output decoding of convolutional
codes. Section 5 presents the very first hardware turbo codecs, some pioneer telecom-
munication applications having adopted these codes, and an overall picture of the
current telecommunication standards including turbo codes. Section 6 concludes the
chapter.

2 History of turbo codes

The invention of turbo codes is the outcome of an intuitive experimental approach,
inspired by the work of some European researchers, Gerard Battail, Joachim
Hagenauer, and Peter Hoeher who, at the end of the 1980s [4–7], highlighted the
interest of soft-output decoding for concatenated coding. This section presents a
1 http://www.telecom-bretagne.eu/index.php?lang=en_GB.
 2 History of Turbo Codes 3

chronology describing the successive ideas that led to the development of the first
turbo codes, whose publication in 1993 [8] shook the coding community. With a
performance at 0.5 dB from the Shannon limit, they showed a gain of almost 3 dB
compared to solutions existing at that time.

2.1 The origins of turbo codes

The origins of turbo codes are to be found in the late 1980s at Telecom Bretagne.
Every year, a group of third year students directed by Claude Berrou and Patrick
Adde was assigned to implement a digital communication function into a CMOS
logic integrated circuit. In 1989, Alain Glavieux suggested the Soft-Output Viterbi
Algorithm (SOVA) proposed by Battail in [4] for implementation. Therefore, the
beginning of the work on error-correcting codes at Telecom Bretagne was marked by
essential references on Viterbi decoding such as [9,10] and by the two main papers
describing modifications to the Viterbi decoder to make it provide soft decisions,
[4,7]. After two years, a suitable hardware architecture was finally proposed [11].
Meanwhile, this work led the researchers to a deep understanding of probabilistic
decoding. Following Battail, Hagenauer, and Hoeher, it was observed that a soft-input
and soft-output (SISO) decoder could be regarded as a signal-to-noise ratio (SNR)
amplifier. This observation stimulated Berrou to consider techniques commonly used
with electronic amplifiers, especially negative feedback. However, this analogy with
amplifiers is meaningful only if the input and output of the decoder represent the same
signal—that is to say, only if the code is systematic.
The next phases went faster. The development of turbo codes passed through
several stages and led to the introduction of neologisms, such as parallel concatenation
and extrinsic information, now part of the coding theory jargon. Here in a few words
are the reflections that marked out this work, as related in [12].

2.2 Concatenation
Using the version of the SOVA in [11], it was possible to cascade SNR amplifiers
and do the experiments described in [6], which was the initial plan: decoding a
classical—i.e., serial—concatenation of two or more ordinary—i.e., non-systematic,
non-recursive—convolutional codes. Concatenation is a simple way to obtain large
asymptotic gains [13], but the performance at low SNR is debased due to the sharing
of the redundancy energy between the component codes.
The concatenated coding and decoding scheme that served as a starting point to
develop turbo codes is described in Figure 1.

2.3 Negative feedback in the decoder and recursive

systematic convolutional codes
Analyzing the scheme in Figure 1, one can notice a dissymmetry in the use of received
information: the inner decoder benefits only from redundancy symbols y1 whereas the
outer decoder benefits from redundancy symbols y2 and from the work of the inner
4 CHAPTER 1 Turbo Codes: From First Principles to Recent Standards

Outer Code (Code C2) X2 Inner Code (Code C1)

X1

Parallel Inter- Parallel Output

Input D D D to serial leaver D D D to serial
d
Equivalent
Y2 Puncturing Y1 Puncturing discrete
channel

Noise

x2 Serial to x1 Serial to
parallel Decoder parallel
Output Decoder Deinter-
+ leaver DEC1 +
DEC2 depunct- (SOVA) depunct-
d̂
y2 uring y1 uring

FIGURE 1
Serial (conventional) concatenation of two convolutional codes with coding rates 3/4
(outer code) and 2/3 (inner code). Global coding rate is 1/2.

decoder. This observation gave Berrou the idea of re-injecting the result of outer
decoding into the inner decoder. As the different levels of the composite decoder
do not represent the same pieces of information—the codes are not systematic, it
was necessary to build an estimate of symbols x2 and y2 at the output of decoder
DEC2. At first, it was a great surprise to observe that the bit error rate (BER) of these
reconstructed symbols after decoding was lower than the BER of decoded information
d̂. However, an intense search did not make it possible to find any explanation for
this strange behavior in the literature. It was then a straightforward task to (re)invent
recursive systematic convolutional (RSC) codes in order to have information data
carried by the encoder output instead of its state and take advantage of this property
not covered in other works. An insight into the distance properties of these codes
is given in Section 3.1. The detailed analysis can be found in [14] (in French). The
resulting serial concatenated coding scheme is shown in Figure 2.
The idea of re-injecting the result of outer decoding into the inner decoder, sim-
ilar to the principle of the turbo engine, gave its prefix to turbo codes, although it
would have been more rigorous to mention only turbo decoding, since no feedback
is implemented at the encoder side.

2.4 Extrinsic information and iterative decoding

The soft-output Viterbi decoder of a systematic code provides a good estimate of
the log likelihood ratio (LLR) relative to its input symbols. It can be shown that
each computed LLR can be expressed as the sum of two contributions. The first,
intrinsic information, is available at the channel output before any decoding stage;
 2 History of Turbo Codes 5

Outer Code (RSC) X2=d Inner Code (RSC) X1

Parallel Inter- Parallel Output

D D D to serial leaver D D D to serial
Input
d

Y2 Puncturing Y1 Puncturing

FIGURE 2
Serial (conventional) concatenation of two recursive systematic convolutional (RSC) codes
with coding rates 3/4 (outer code) and 2/3 (inner code). Global coding rate is 1/2.

LLR (x2)
+
- Parallel Inter- z
LLR (y2) to leaver
+ serial Extrinsic
- information

Serial to x1 Serial to
Decoder parallel Decoder parallel From
x2
-
DEC2 + Deinter- +
leaver
+ DEC1
Output depunct- depunct- channel
(SOVA) (SOVA)
y2 uring y1 uring
dˆ
LLR(x1)
LLR(d)

FIGURE 3
Structure of the very first turbo decoding scheme.

the second, extrinsic information, is provided by exploiting the dependencies (due to

convolution, parity, …) existing between the symbol being processed and the other
symbols processed by the decoder. As the intrinsic information is used by both
decoders (at different instants), the extrinsic information produced by each of the
decoders must be passed to the other as new information to ensure joint convergence.
In any decoding construction, extrinsic information must not be used by the processor
which produced it. Berrou finally came up with the scheme depicted in Figure 3. The
decoder is built in such a way that no information produced by a component decoder
can feed its input, either directly or indirectly.
Others, mainly in the United States, Elias [15], Gallager [16,17], Tanner [18],
and so on had earlier imagined procedures for coding and decoding that were the
forerunners of turbo codes.
The digital processing of information has at least one major drawback: it does
not lend itself easily to feedback techniques. An iterative procedure has to be used
because of the delays (trellis, interleaving, …). This procedure increases the decoder
latency, but ever-continuing progress in microelectronics makes possible today what
was unimaginable not so long ago.
6 CHAPTER 1 Turbo Codes: From First Principles to Recent Standards

Two hardware solutions can be contemplated, depending on the information rate.

For low data rates, a single decoding processor working at a high data frequency
can make all the required iterations with a tolerable added delay. For high rates, a
cascade of identical decoding modules can be implemented monolithically to enable
high-speed pipeline processing (in this case, typically that of broadcasting, the latency
problems are generally less crucial).

2.5 Parallel concatenation

The idea of the so-called parallel concatenation paradigm did not germinate by anal-
ogy with product codes, as sometimes reported. It came actually from the team in
charge of the design of the very first pipelined turbo decoding integrated circuit,
intended to validate the concept of iterative decoding. Serial concatenation requires
different clock signals for the inner and outer decoders. Due to the symmetric structure
of the encoder, parallel concatenation simplifies the architecture of the system since
both component encoders and decoders can work with the same clock signal, which
is also the data clock. The first studied parallel turbo encoder and its decoding scheme
are shown in Figure 4. Turbo codes are sometimes also called parallel concatenated
convolutional codes (PCCCs) in contrast with the conventional serial concatenation.

First Code (RSC) Second Code (RSC) X1

Inter- Parallel Output

Input D D D leaver D D D to serial
d
Equivalent
Y1 Puncturing Y2 Puncturing discrete
channel

Noise +

+ Deinter- z
LLR2(d) - leaver Extrinsic
information

x1 Serial to
Output Decoder - Decoder parallel
DEC2 Inter- + DEC1 +
leaver y1 depunct-
dˆ (SOVA2) (SOVA1)
y2 uring
LLR1(d)

FIGURE 4
Parallel concatenation of two RSC codes and associated (asymmetrical) decoder.
 2 History of Turbo Codes 7

+ Inter-
LLR1(d) - leaver

x1
Decoder
DEC1
z2
(SOVA1) y1
Serial to
parallel From
+
y2 depunct- channel
Decoder uring
Output
DEC2 z1
dˆ x2 Inter-
(SOVA2)
leaver

Deinter-
-
LLR2(d) +
leaver

FIGURE 5
Symmetrical turbo decoder.

Later on, a symmetrical structure was also devised for the turbo decoder
(Figure 5), which is more natural is the sense that it reflects the symmetry of the
encoding process.
It was also observed that, for a given coding rate, parallel concatenation yields
more redundant symbols from the outer code than serial concatenation does. A parallel
concatenation of two elementary codes C1 and C2 with coding rates R1 and R2 has
a global coding rate:

R1 R2 R1 R2
Rp = = , (1)
R1 + R2 − R1 R2 1 − (1 − R1 )(1 − R2 )

whereas the global coding rate of the serially concatenated code is R S = R1 R2 . For
instance, a global coding rate 1/2 can be obtained with the parallel concatenation of
two codes with elementary rates 2/3 or with the serial concatenation of two codes
with elementary rates 2/3 and 3/4, as in Figure 1. Thanks to a greater number of
redundant symbols, the parallel structure can benefit from a higher diversity effect.
This explains why the convergence threshold, i.e., the minimum SNR at which the
iterative decoder starts to correct most of the errors, is lower when the concatenation
is parallel. In return, serially concatenated convolutional codes (SCCCs) show lower
changes of slope in the bit error probability curves than their parallel counterpart,
due to higher minimum Hamming distances. The distance properties of PCCCs and
SCCCs are analyzed in [19,20].
8 CHAPTER 1 Turbo Codes: From First Principles to Recent Standards

3 Fundamentals of turbo coding

The turbo encoder involves a parallel concatenation of at least two elementary recur-
sive systematic convolutional (RSC) codes separated by an interleaver ().
Figure 6 represents a turbo code in its most conventional version [8,21]. The
input binary message of length K is encoded in its natural order and in a permuted—
or interleaved—order by two RSC encoders ENC1 and ENC2. In the figure, the RSC
encoders are identical but this property is not essential. The overall coding rate R of
the turbo code is 1/3. To obtain higher rates, the most common technique involves
puncturing, i.e., not transmitting, part of coded symbols, most often redundancy sym-
bols Y1 and Y2. Adopting m-binary codes is another way of increasing the code
rate [22].

3.1 Recursive systematic convolutional (RSC) component codes

The convolutional codes used in the experiments described in Figure 1 were a non-
systematic code. However, due to the observation mentioned in Section 2.3 that the
BER of the reconstructed symbols after decoding was lower than the BER of decoded
information, Berrou, Glavieux, and their PhD student Thitimajshima investigated the
ways of building systematic convolutional codes with good performance [14]. It was
then observed that recursive systematic codes could offer good performance both

ENC1

Data
D D D

Y1
Interleaver
ENC2

Possible
puncturing
D D D

Y2

FIGURE 6
The turbo encoder structure, a parallel concatenation of two RSC encoders separated by
an interleaver.
Exploring the Variety of Random
Documents with Different Content
to make
Clary Wine.
Take twenty-four pounds of Malaga raisins, pick and chop them very small, put
them in a tub, and to each pound a quart of water; let them steep ten or eleven
days—this sounds like a school treat—stirring it twice every day; you must keep it
covered close all the while; then strain it off, and put it into a vessel, and about
half a peck of the tops of clary (what was clary?) when ’tis in blossom; stop it
close for six weeks, and then bottle it off; in two or three months ’tis fit to drink.

Clary naturally leads to

Apricock Wine,
which we of the nineteenth century miscall apricot. The derivation of
the word is Latin. Then the Arabs got hold of it, and it became Al-
precoc. Then the thriving Spaniards got hold of the word, which
became Alborcoque; and so to England. But to the wine. {35}

Take three pounds of sugar, and three quarts of water, let them boil together,
and scum it well; then put in six pounds of apricocks, pared and stoned, and let
them boil till they are tender; then take them up, and when the liquor is cold
bottle it up. You may, if you please, after you have taken out the apricocks, let the
liquor have one boil with a sprig of flower’d clary in it.

Also, you may if you please—and you probably will please—add a

little old brandy to the decoction.
 CHAPTER IV
SOME OLD RECIPES
Indifference of the Chineses — A nasty potion — A nastier — White Bastard —
Helping it to be eager — Improving Malmsey — Death of the Duke of Clarence —
Mum is not the word — English champagne — Life without Ebulum a blank — Cock
ale — How to dispose of surplus poultry — Painful fate of a pauper — Potage pauvre
— Duties of the old English housewife — Election of wines, not golf — Muskadine —
Lemon wine — Familiar recipe — King William’s posset — Pope’s ditto.

“The Chineses,” says a very old work on liquid nourish­ment, “make

ex­cel­lent Drink of Rice, which is very pleasant of taste, and is pre­‐
ferred by them before wine.”
But, like the Germans, the Chineses will eat and drink pretty
nearly anything. And this is the cheering mixture which the Chineses
sampled in the new German colony of Kiant-schan, according to the
Frankfurter Zeitung:—
“Sitting under the poplars one can imagine oneself in the
courtyard of an old German feudal castle. The hamper is opened,
and the cold mountain stream flowing before the temple serves as
an ice cellar. Once more the male population of the village puts in an
appearance, standing round the table in amazement at all the {37}
unheard-of things happening. The greatest success attends the
uncorking of the Apollinaris bottles. The bottle is pointed at the
onlookers, and the cork having been loosened it flies into their faces
with a loud report. At first they are greatly alarmed, then they enjoy
the joke hugely, and at last they all squat on the ground in a circle,
and send a deputy to the table of the foreigners, bearing a teacup.
The petition is granted, and in the teacup an exquisite brew is
prepared. The drainings of all the beer bottles are collected, to
which is added a little claret and a liberal proportion of Apollinaris,
and then, in order to lend greater consistency to the beverage, some
sausage skins are mixed with it. The teacup circulates amongst the
Chinese, and each sips it with reverential awe. Some of them make
fearful grimaces, but not one has the courage of his opinion, and it
is evident that, on the whole, the drink is voted a good one,
although, perhaps, its flavour is somewhat rare.”
Next, please. Oh, here is another, about some neighbours of the
Chineses.
“In the Isle Formosa, not far from China, the Natives make a
Drink as strong and intoxicative as Sack, out of Rice, which they
soak in warm water, and then beat it to a paste in a Mortar; then
they chew some Rice-meal in their mouths, which they spit to a pot
till they have got about a quart of liquor, which they put to the paste
instead of Leaven or Ferment. And after all be kneaded together till
it be Dough, they put it into a great earthen pot, which they fill {38}
up with water, and so let it remain for two months; by which means
they make one of the most pleasant Liquors a man need drink; the
older the better and sweeter, although you keep it five and twenty or
thirty years.”
Weel—I hae ma doots.
Until reading “The English Housewife, containing the inward and
outward Vertues which ought to be in a complete Woman, published
by Nicholas Okes at the sign of the golden Unicorne, in 1631,” I had
no skill in making
White Bastard
or “aparelling” Muskadine. They used a lot of eggs in the vintry in
those days, and these were the instructions for making white
bastard.
Draw out of a pipe of bastard ten gallans, and put to it five gallans of new
milke, and skim it as before, and all to beat it with a parill of eight whites of
egges, and a handfull of Baysalt and a pint of conduit-water, and it will be white
and fine in the morning. But if you will make very fine bastard—which I,
personally, have no ambition to do—take a white-wine hog’s-head, and put out the
lees, and wash it cleane, and fill it halfe full and halfe a quarter, and put to it foure
gallans of new milke, and beate it well with the whites of sixe egges, and fill it up
with white-wine and sacke, and it will be white and fine.

Bastard had not much rest in the seventeenth century. The

housewife who might wish “to helpe bastard being eager” had to
follow these directions:—
Take two gallons of the best stoned honey, and two gallons of white wine, {39}

and boyle them in a faire panne, skimme it cleane, and straine it through a
faire cloth that there be no moats in it; then put to it one ounce of collianders
(coriander seeds?) and one ounce of aniseeds, foure or five orange pils (pips?) dry
and beaten to powder, let them lye three dayes; then draw your bastard into a
cleane pipe, then put in your honey with the rest, and beate it well; then let it lye
a weeke and touch it not, after draw it at pleasure.

In the present enlightened century such a recipe does not read

like helping the possible consumer to be “eager.”
Nor does the following method of treating Malmsey sound
promising, except for making its consumer particularly “for’ard”:—
If you have a good but of Malmsey, and a but or two of sacke that will not be
drunke; for the sacke prepare some empty but or pipe, and draw it more than
halfe full of sacke; then fill it up with Malmsey, and when your but is full within a
little, put into it three gallons of Spanish cute, the best that you can get—where
did they get it?—then beate it well; then take your taster, and see that it bee
deepe coloured; then fill it up with sacke, and give it aparell, and beate it well.
The aparell is thus: Take the yelkes of tenne egges and beate them in a cleane
bason with a handful of Bay salt, and a quarte of conduit-water, and beate them
together with a little peece of birch, and beate it till it be as short as mosse; then
draw five or sixe gallons out of your but, then beate it againe, and then fill it up,
and the next day it will be ready to be drawne. This aparell will serve both for
muscadine, bastard, and for sacke. {40}

We are not told in history if the butt of Malmsey in which the

Duke of Clarence shuffled off his mortal and sinful coil had been
previously subjected to this “aparell” and castigation. In the interests
of mercy, let us hope not.
The fluid once known as
Mum
never claimed any sort of relationship with sparkling wine, but was a
species of unsophisticated ale, brewed from wheat, or oats, with a
little bean-meal occasionally introduced; in fact, the sort of stuff we
use in the present century to fatten bacon pigs upon. And “mum”
has not been the word with British brewers for some time past.
Champagne has been made in England for a considerable period;
but since the closing of the “night-houses” in Panton Street the trade
therein has not been very brisk. During the present century
champagne in this country—and I grieve to add in France as well—
has been chiefly fabricated from apples, and other fruits; but here is
a much older way of making
English Champagne.
Take to three gallons of water nine pounds of Lisbon sugar; boil the water and
sugar half an hour, scum it clean, then have one gallon of currants pick’d, but not
bruis’d, pour the liquor boiling hot over them, and when cold work it with half a
pint of balm two days; then pour it through a flannel or sieve, then put it into a
barrel fit for it with half an ounce of ising-glass well bruis’d. When it has done
working stop it close for a month, then bottle it, and in every bottle put a very {41}

small lump of double-refin’d sugar. This is excellent wine, and has a beautiful
colour.

“Life without Ebulum,” writes a friend, an instructor of youth in

the ingenuous arts, in forwarding me the recipe, “is a void to most
people who have not cultivated the eringo root in their back
gardens.” I have never tasted ebulum, preferring my ale neat and
unadorned, but this is how to prepare
Ebulum.
To a hogshead of strong ale take a heap’d bushel of elderberries, and half a
pound of juniper berries beaten; put in all the berries when you put in the hops,
and let them boil together till the berries break in pieces; then work it up as you
do ale. When it has done working, add to it half a pound of ginger, half an ounce
of cloves, as much mace, an ounce of nutmegs, and as much cinnamon grosly
beaten, half a pound of citron, as much eringo root, and likewise of candied
orange-peel. Let the sweetmeats be cut in pieces very thin, and put with the spice
into a bag, and hang it in the vessel when you stop it up. So let it stand till ’tis
fine, then bottle it up, and drink it with lumps of double-refin’d sugar in the glass.
 One of the quaintest beverages of which I ever heard, or read, is
Cock Ale.
In order to make this, the Compleat Housewyfe instructs us to take ten gallons
of ale, and a large cock, the older the better. Parboil the cock, flea (flay?) him, {42}

and stamp him in a stone mortar till his bones are broken (you must craw and
gut him when you flea him), then put the cock into two quarts of sack, and put to
it three pounds of raisins of the sun stoned, some blades of mace, and a few
cloves; put all these into a canvas bag, and a little before you find the ale has
done working, put the ale and bag together into a vessel; in a week or nine days
bottle it up, fill the bottles but just above the neck, and give it the same time to
ripen as other ale.

I have frequently read of the giving of “body” to ale and stout, by

means of the introduction of horseflesh; and an old song used to tell
us that upon one of the paupers in a certain workhouse happening,
in­ad­ver­tently, to fall head-fore­most into the copper,
dreadful to tell, he was boiled in the soup,
which, on that account, in all probability so strengthened the con­sti­‐
tu­tions of the other paupers as to render them impatient of work­‐
house dis­ci­pline. The man who disap­peared mysteriously—this is Mr.
Samuel Weller’s story—and who unwittingly furnished “body” for the
sausages supplied to the neigh­bour­hood, was, after all, benefiting
his fellow-men. But to put the rooster into the ale-cask smacks
somewhat of barbarism; and thank goodness we do not work off our
surplus poultry in that fashion nowadays. But these barbarians were
not ashamed; for lo! facing me is “another way” for the man­u­fac­ture
of rooster-beer.
Take an old red, or other cock, and boyle him indifferent well; then flea his {43}

skin clean off, and beat him flesh and bones in a stone mortar all to mash,
then slice into him half a pound of dates, two nutmegs quartered, two or three
blaids of mace, four cloves; and put to all this two quarts of sack that is very
good; stop all this up very close that no air may get to it for the space of sixteen
hours; then tun eight gallons of strong ale into your barrel so timely as it may
have done working at the sixteen hours’ end; and then put thereinto your infusion
and stop it close for five days, then bottle it in stone bottles; be sure your corks
are very good, and tye them with pack-thread; and about a fortnight or three
weeks after you may begin to drink of it; you must also put into your infusion two
pound of raisins of the sun stoned.

Holy Moses ! What a drink !

“It is necessary,” wrote a chronicler of the day, “that our English
Housewife be skilfull in the election, preservation, and curing of all
sorts of wines, because they be usuall charges under her hands, and
by the least neglect must turne the Husband to much losse.”
This was written, I may interpolate, before the bicycle craze had
set in, and before the era of ladies’ clubs. Fancy asking the New
Woman to elect, preserve, and cure all sorts of wines!
“Therefore,” continues the same writer, “to speak first of the
election of sweete Wines she must be careful that her Malmseys be
full Wines, pleasant, well hewed, and fine; that Bastard be fat, and if
it be tawny it skils not, for the tawny Bastards be always the
sweetest. Muskadine must be great, pleasant, and strong, with a
swete sent, and Amber colour. Sacke, if it be Seres (Xerez?), {44}
which it should be, you shall know it by the marke of a corke burned
on one side of the bung, and they be ever full gadge, and so are no
other Sackes; and the longer they lye the better they be.”
Muskadine
was, apparently, made from bastard and malmsey, with the addition
of ginger and new milk (with the cream removed).
 Here is a potion bearing the harmless, Band-of-Hopish name of
Lemon Wine,
which would not, however, be tolerated at a Salvation Army banquet.
The first part of the recipe will be familiar to many of my young
friends.
Take six large lemons, pare off the rind, and cut the lemons and squeeze out
the juice, and in the juice steep the rind, and put it to a quart of brandy—so far,
brother, the court is with you—and let it stand in an earthen pot close stop’t three
days, and then squeeze six more, and mix with two quarts of spring-water, and as
much sugar as will sweeten the whole, and boil the water and lemons and sugar
together, and let it stand till ’tis cool. Then add a quart of white wine and the other
lemon and brandy, and mix them together, and run it through a flannel bag into
some vessel. Let it stand three months and bottle it off. Cork your bottles very
well, and keep it cool; it will be fit to drink in a month or six weeks. {45}

Cheer-oh ! This potion reads well, and I know a punch which

bears some resemblance thereto. But why call it lemon wine? Do not
the brandy and the white wine deserve some recognition in the
nomenclature?
What is understood by the name
Barley Wine
nowadays is a particularly strong brew of ale. With the ancients,
however, it was a drink which might have been with safety handed
round at breaking-up parties in a young ladies’ school.
Take half a pound of French barley, and boil it in three waters, and save three
pints of the last water, and mix it with a quart of white wine, half a pint of borage-
water, as much clary-water, and a little red rose-water, the juice of five or six
lemons, three quarters of a pound of fine sugar, and the thin yellow rind of a
lemon; brew all these quick together, run it through a strainer, and bottle it up. ’Tis
pleasant in hot weather, and very good in fevers.
 In the matter of possets—of which more anon—the following
reads like a seductive winter’s beverage, especially if the imbiber
have a cold in the head. Fear not the bile, but read the directions for
making
King William’s Posset.
Take a quart of cream, and mix with it a pint of ale, then beat the yolks of ten
eggs and the whites of four; when they are well beaten, put them to your {46}

cream and ale. Sweeten to your taste and slice some nutmeg in it; set it over
the fire, and keep it stirring all the while, and when ’tis thick, and before it boils,
take it off, and pour it into the bason you serve it in to the table.

Here is another, even more seductive.

To make the Pope’s Posset.
Blanch and beat three-quarters of a pound of almonds so fine that they will
spread between your fingers like butter, put in water as you beat them to keep
them from oiling. Then take a pint of sack or sherry, and sweeten it very well in
double-refin’d sugar, make it boiling hot, and at the same time put half a pint of
water to your almonds, and make them boil; then take both off the fire, and mix
them very well together with a spoon. Serve it in a china dish.

Frontiniac Wine
was simplicity itself.
Take six gallons of water and twelve pounds of white sugar, and six pounds of
raisins of the sun cut small; boil them together one hour; then take of the flowers
of elder, when they are falling and will shake off, the quantity of half a peck; put
them in the liquor when ’tis almost cold, and next day put in six spoonfuls of syrup
of lemons, and four spoonfuls of ale yeast; and two days after put it into a vessel
that is fit for it, and when it has stood two months bottle it off.
 In the olden times, just before Oliver Cromwell was a going
concern, there were two sorts of what was then called {47}

Renish Wine,
that is to say, Elstertune and Barabant.
“The Elstertune,” says my informant, “are best, you shall know it
by the Fat, for it is double bard and double pinned”—I have not the
faintest idea what he means, but those are his words; “the Barabant
is nothing so good, and there is not so much good to be done with
them as with the other. If the Wines be good and pleasant, a man
may rid away a Hogshead or two of White wine, and this is the most
vantage a man can have by them; and if it be slender and hard,
then take three or four gallons of stone-honey and clarify it cleane;
then put into the honey four or five gallons of the same wine, and
then let it seeth a great while, and put into it twopence in cloves
bruised, let them seeth together, for it will take away the sent of
honey; and when it is sodden take it off, and set it by till it be
thorow cold; then take foure gallons of milke and order it as before,
and then put all into your wine, and all to beate it; and (if you can)
role it, for that is the best way; then stop it close and let it lie, and
that will make it pleasant.”
Possibly, but it seems a deal of trouble to take over a wine.
And now let us adjourn to a more familiar subject, for discussion
in the next chapter.
 CHAPTER V
GLORIOUS BEER
Nectar on Olympus — Beer and the Bible — “Ninepenny” at Eton — “Number One”
Bass — “The wicked weed called hops” — All is not beer that’s bitter — Pathetic story
of “Poor Richard” — Secrets of brewing — Gervase Markham — An “espen” full of
hops — Eggs in ale — Beer soup — The wassail bowl — Sir Watkin Wynne — Brown
Betty — Rumfustian — Mother-in-law — A delightful summer drink — Brasenose ale.

As much poetry has been written in praise of John Bar­ley­corn as in

praise of wine, woman, battles, heroes, Cupid’s darts, and patent
med­ic­ ines. And one dear old song, which seems to ring in my ears
as I write, proclaimed that in the opinion of the author the nectar
which the gods imbibed from golden goblets on the top of Mount
Olympus was in reality cool, re­fresh­ing pale ale, quaffed out of
pewter tankards. Whether this was so matters not, but as to the
antiquity of beer as a beverage there can be no question; and
however much the demand for other liquors may have slackened
during the rolling on of time, John Barleycorn is still growing in
public estimation. Breweries keep on springing up all over the {49}
country, and those who purchase shares in them receive, for the
most part, sub­stan­tial div­id
­ ends. “Beer and the Bible” have won
more elections than any other combination; the organ­iz­ a­tion of the
brewers has hitherto proved powerful enough to withstand all the
slings and arrows of the Prohibition party, whilst there has been an
enormous increase in the value of houses licensed to sell fermented
refreshment; and the name of Bass will “live on,” like Claudian,
“through the centuries.”
 There be more than one description of beer put before the
public. I forget at this moment who was responsible for the “swipes”
of my school days, which tasted like red ink—and I have sampled
both—but I have always believed that the man­u­fac­turer—I do not
believe him to have been a brewer at all—had a special spite against
the rising generation, which he wished to die a lingering death. The
“ninepenny” quaffed beneath the holy shade of Henry was good,
sound, wholesome tipple; but I fancy an inferior brand was poured
forth to us at “half time” in the football field. Since those days I have
tasted pretty nearly all sorts and conditions of beer, from the
“Number One” Bass drawn from the wood in pewter pots, in a little
hostelry just off the Waterloo Road—the very best according to my
taste—to the awful stuff tasted, and only tasted, one Sunday in a
charmingly rural-looking little inn, with a thatched roof—a licensed
house which apparently laid itself out to entrap the daring and
enterprising “bona fide traveller,” and whose malt liquor was
apparently composed for the most part of vinegar and dirty {50}
water, in which had been soaked quassia chips, salt, bloater-heads,
and some of the thatch from the roof.
Beer was the current name in England for every description of
malt liquor before the introduction of “the wicked weed called hops”
from the Netherlands in 1524. According to the Alvismal, a didactic
Scandinavian poem of the tenth century, this malt liquor was called
ale amongst men, and beer by the gods; and it was probably from
this Scandinavian poem that the author of the anything-but-didactic
poem quoted above got his ideas as to the real nature of the
beverage partaken of on Olympus. In the Eastern counties of
England, and over the greater part of the kingdom, ale signifies
strong, and beer small, malt liquor, but in the West these names
mean exactly the reverse—which must be confusing in the extreme
to the intelligent foreigner on his travels in search of facts and—
refreshment. As now used, ale is distinguished from beer—I am
alluding to the more civilized parts of our country—chiefly by its
strength, and by the quantity of sugar remaining in it
undecomposed. Strong ale is made from the best pale malt, and the
fermentation is allowed to proceed slowly, and the ferment to be
exhausted and separated. This, together with the large quantity of
sugar still left undecomposed, enables the liquor to keep long,
without requiring a large amount of hops.
The last few lines may give the reader the impression that the
writer served his time in Burton-on-Trent; but this is not the case. I
have conveyed the bulk of my technical knowledge of brewing {51}
from standard works on the subject.
It will be gathered from some previous remarks that all is not
beer that’s bitter; and although it would seem impossible to find a
cleaner, healthier, or more strengthening drink than the “pure beer”
of commerce, brewed from good English or Scotch barley, Kentish
hops, and fair spring-water, how about the wash sold in some
licensed houses which is “fetched up” with foot-sugar, bittered with
quassia, and mixed with salt and any nasty flavourer which is handy?
The old stories about the carcass of a horse placed in the London
stout, to give it “body,” and the mysterious disappearance of an
Italian organ-grinder, together with his monkey and infernal
machine, just outside a high-class brewery, are probably apocryphal.
And although the ancients undoubtedly put a red cock—the older the
better—into ale, on occasion, the nineteenth century Briton, for the
most part, if the rooster be too tough to serve as a boiled bonne
bouche with parsley-and-butter, usually makes Cock-a-Leekie of him.
And thereby hangs a tale.
When my firm was running a small chicken-ranche we once
reared an unfortunate fowl, who had curvature of the spine, almost
from the fracture of his shell. He was a weakling, and his brethren
and sistren, after the manner of birds, beasts, and fishes, who “go
for” the anæmic and infirm, persecuted him exceedingly, and peeked
most of his feathers off. Being a merciful, and withal a thrifty, {52}
poultry-farmer, I looked out an old parrot’s cage from the tool-shed,
and in this cage installed the weakly cockerel. He was forthwith
christened “Poor Richard,” and given little Benjamin’s share of the
corn and wine, and cayenne pepper and—other things. And although
his head was still slewed round to starboard, he thrived under his
liberal nourishment and freedom from the assaults of his relatives.
Time flew on. I had been the “Northern Circuit,” in the pursuit of
my then profession of reporter of the sport of kings. I returned
home late on a Saturday night, and next day we had friends to
dinner. So much North Country language, and so much travelling
about had quite put our feathered and afflicted pensioner out of my
head; and even the fact of our having the favourite broth of His
Majesty King James the First for dinner did not suggest anything to
my busy brain. But afterwards, when we were alone—she ought not
to have done it—my life-partner confided to me that I had helped to
eat “Poor Richard”! And I felt like a very cannibal; and mourned the
bird as a brother.
 But to return. In Queen Elizabeth’s reign it was, I used to believe,
a capital offence to put hops into beer. But these are the directions
for
Brewing of Strong Ale,
issued by one Gervase Markham, an authority on the subject, and a
contemporary of Shakespeare; and in these directions “hops” are
distinctly mentioned as one of the component parts of the brew. {53}
Now for the brewing of strong Ale, because it is drinke of no such long lasting
as Beere is, therefore you shall brew lesse quantity at a time thereof, as two
bushels of Northerne measure (which is foure bushels or halfe a quarter in the
South) at a brewing, and not above, which will make foureteene gallons of the
best Ale. Now for the mashing and ordering of it in the mash-fat, it will not differ
any thing from that of Beere; as for hops, although some use not to put in any,
yet the best Brewers thereof will allow to foureteene gallons of Ale a good espen
full of hops, and no more, yet before you put in your hops, as soone as you take it
from the graines, you shall put it into a vessell, and change it, or blinke it in this
manner: Put into the Wort a handfull of Oke-bowes and a pewter dish, and let
them lye therein till the Wort looke a little paler than it did at the first, and then
presently take out the dish and the leafe, and then boile it a full houre with the
hops, as aforesayd, and then clense it, and set it in vessels to cook; when it is
milk-warme, having set your Barme to rise with some sweete Wort; then put all
into the guilfat, and as soone as it riseth, with a dishe or bowle beate it in, and so
keepe it with continuall beating a day and a night, and after run it. From this Ale
you may also draw halfe so much very good middle Ale, and a third part very good
small Ale.

Another way
To make Strong Beer
was published at a later date than the above, and to my thinking is
not a better way.
 To a barrel of beer take two bushels of malt and half a bushel of wheat just
crackt in the mill, and some of the flour lifted out of it; when your water is {54}

scalding hot, put it in your mashing-fat; there let it stand till you can see your
face in it; then put your malt upon it, then put your wheat upon that, and do not
stir it; let it stand two hours and a half; then let it run into a tub that has two
pounds of hops in it, and a handful of rosemary flowers, and when ’tis all run put
it in your copper and boil it two hours; then strain it off, setting it a-cooling very
thin, and set it a-working very cool; clear it very well before you put it a-working,
put a little yeast to it; when the yeast begins to fall, put it into your vessel, and
when it has done working in the vessel, put in a pint of whole wheat and six eggs;
then stop it up, let it stand a year, and then bottle it. Then mash again, stir the
malt very well in, and let it stand two hours, and let that run, and mash again, and
stir it as before; be sure you cover your mashing-fat well up, mix the first and
second running together; it will make good household beer.

I rather fancy the blend­ing of a lot of eggs (pre­sum­ably new-laid)

with the mash, would “break” some of the smal­ler brewers. It could
hardly be done at the price.
The Germans make
Beer Soup.
Whether this is made from British or lager beer is not stated in the
recipe before me, which hardly reads suited to the ordinary English
palate.
I will now give a few modern recipes for tasty beer-compounds.
{55}
Ale Cup (Cold).
Squeeze the juice of a lemon into a round of hot toast; lay on it a thin piece of
the rind, a tablespoonful of powdered sugar, a little grated nutmeg or powdered
all-spice, and a sprig of balm. Pour over these one wine-glass of brandy, two of
sherry, and three pints of mild ale. Do not allow the balm to remain in many
minutes.

Ale Flip (Hot).

 Put into a saucepan three pints of ale, a tablespoonful of sugar, a blade of
mace, a clove, and a small piece of butter, and bring the liquor to a boil. Beat the
white of one egg and the yolks of two thoroughly, mixing with them a
tablespoonful of cold ale. Mix all together, and then pour the whole rapidly from
one large jug to another, from a good height—mind your fingers and the furniture
—for some minutes, to froth it thoroughly. Do not allow it to get cool.

Ale Posset (Hot).

Boil a pint of new milk, and pour it over a slice of toasted bread. Stir in the
beaten yolk of an egg and a small piece of butter, and sugar ad lib. Mix these with
a pint of hot ale, and boil for a few minutes. When the scum rises the mixture is
ready for use.

Mulled Ale (Very Hot).

Put half a pint of ale, a clove, a little whole ginger, a piece of butter the size of
a marble, and a teaspoonful of sugar into a saucepan, and bring it to boiling- {56}

point. Beat two eggs with a tablespoonful of cold ale, and pour the boiling ale
into them, and then into a large jug. Pass the whole from one jug to another, as in
the case of Ale Flip, return to saucepan, and heat it again till almost, not quite, at
boiling-point.

With regard to
Wassail, or Swig (Cold),
which used to be a very popular beverage at the uni­vers­i­ties—at one
time it was peculiar to Jesus College, Oxford—is of very ancient date
indeed.
“Sir quod he,” is part of a conversation culled from an old MS.,
“Watsayll, for never days of your lyf ne dronk ye of such a cuppe,”
which sounds as if the Watsayll was of a seductive and harmful
nature. Nev­er­the­less here is the recipe, taken from “Oxford
Nightcaps.”
Put into a bowl half a pound of Lisbon sugar (if you do not possess that brand,
I have no doubt “best lump,” pulverized, will do as well), and pour on it one pint of
warm beer; grate a nutmeg and some ginger into it; add four glasses of sherry
and five additional pints of beer; stir it well and sweeten to taste; let it stand
covered up two or three hours, then put three or four slices of bread cut thin and
toasted brown into it, and it is fit for use. Sometimes two or three slices of lemon
are introduced, together with a few lumps of sugar rubbed on the peel of a lemon.
Bottle this mixture, and in a few days it may be drunk in a state of effervescence.

On the festival of St. David, an immense silver-gilt bowl, the gift

of Sir Watkin W. Wynne to the college in 1732 is filled with this {57}
“swig,” and passed round, at Jesus College. And I should prefer to
call the beverage “swig” instead of “wassail,” which should properly
be a hot drink, if we are to believe the illustrated papers at
Christmas-time. And there is no toast in the orthodox Wassail, but,
instead, roasted apples. What does Puck say in A Midsummer Night’s
Dream?
Sometime lurk I in a gossip’s bowl,
In very likeness of a roasted crab,
And when she drinks against her lips I bob,
And on her wither’d dewlaps pour the ale.

Brown Betty
Here is another old recipe:—
Dissolve a quarter of a pound of brown sugar in one pint of water, slice a
lemon into it, let it stand a quarter of an hour, then add a small quantity of
pulverised cloves and cinnamon, half a pint of brandy, and one quart of good
strong ale; stir it well together, put a couple of slices of toasted bread in it, grate
some nutmeg and ginger over the toast, and it is fit for use. Ice it well, and it will
prove a good summer, warm it and it will become a pleasant winter, beverage. It is
drunk chiefly at dinner.

Rather heavily loaded for a dinner drink, I should say.

 Another recipe for
Ale Flip
will serve, here. {58}

Beat well together in a jug, four eggs with a quarter of a pound of sifted sugar;
then add by degrees, stirring all the time, two quarts of old Burton ale, and half a
pint of gin; pour backwards and forwards from one jug to another, and when well
frothed serve in tumblers. Grate a little nutmeg atop of each portion. This is one of
the best “nightcaps” I know—especially after you may have been badger-hunting,
or burgling, or serenading anybody on Christmas Eve.

Rumfustian.
Beat up in a jug, the yolks of two eggs with a tablespoonful of sifted sugar;
then take half a pint of old Burton ale, one wine-glass of gin, one wine-glass of
sherry, a little spice and lemon rind. Let the ale, wine, and gin, mixed together
come to the boil, then pour in the egg mixture, whisking rapidly; serve hot, with a
little nutmeg grated atop.

Such compound drinks, into which ale enters, as Shandy-gaff

require no mention here. Suffice it to mention that this gaff has for
many years been the favourite beverage of those who go up the
river—there is but one river in England—in boats, whether
schoolboys, or of riper years. In Stock Exchange circles champagne
is occasionally substituted for ginger-beer, but this is a combination
in which I have no implicit belief; although champagne and
Guinness’s stout make an excellent mixture. Stout and bitter,
otherwise known as
Mother-in-law,
and old-and-mild, for which the pet name is {59}

Uncle,
are also in much request amongst the groundlings; whilst during the
warm weather I know of no more popular swallow, for moderate
drinkers, who do not require their throats to be scratched, than a
small bottle of lemonade to which is added just one “pull” of pale-
ale. This is called, for the sake of brevity, a
Small Lem and a Dash,
or the Poor Man’s Champagne; and is a refreshing and innocuous
drink which might commend itself to total abstainers.
In the Universities of Oxford and Cambridge there is probably as
much malt liquor drunk per head as in any other part of the world.
Brasenose Ale
has obtained a reputation which the beverage doubtless fully merits.
Since the foundation of this college a custom has prevailed of
introducing into the refectory on Shrove Tuesday, immediately after
dinner, what is denominated Brasenose Ale, but what is known in
many other parts of England as Lamb’s Wool. Verses in praise of the
Ale are—or at all events were—annually written by one of the
undergraduates, and a copy of them is sent to every resident
member of the College.
The following stanzas are taken from one of these
contributions:— {60}
 Shall all our singing now be o’er,
Since Christmas carols fail?
No ! Let us shout one stanza more
In praise of Brasenose Ale !
A fig for Horace and his juice,
Falernian and Massic;
Far better drink can we produce,
Though ’tis not quite so classic.

Not all the liquors Rome e’er had

Can beat our matchless Beer;
Apicius’ self had gone stark mad
To taste such noble cheer.

After all, the potion is simplicity itself:—

Three quarts of ale, sweetened with sifted sugar, and served up in a bowl with
six roasted apples floating in it.
 CHAPTER VI
ALL ALE
Waste not, want not — The right hand for the froth — Arthur Roberts and Phyllis
Broughton — A landlord’s perquisites — Marc Antony and hot coppers — Introduction
of ale into Britain — Burton-on-Trent — Formerly a cotton-spinning centre — A few
statistics — Michael Thomas Bass — A grand old man — Malting barleys — Porter
and stout — Lager beer — Origin of bottled ale — An ancient recipe — Lead-
poisoning — The poor man’s beer.

In a speech made some years ago Sir Michael Hicks-Beach observed

that nearly one million sterling’s worth of tobacco was wasted
annually by throwing away cigarette-ends and the stumps of cigars.
But what would you, Sir Michael? Are the lieges to cremate their lips
and singe their moustaches by smoking on to the (literally) bitter
end? Whether or no, it is tolerably certain that there is an enormous
daily waste in the matter of in­tox­ic­ a­ting drinks—without counting the
wanton, although con­sci­en­tious, des­truc­tion made by teetotal
magnates. According to statistics—I shall not madden my readers
with many of these—more than £138,000,000 are spent annually {62}
in Great Britain on spirituous liquors. Half of this sum, it may be
fairly stated, is spent in the provinces. It may also be taken as read
that 5 per cent of beer and stout is wasted, in the way of froth,
spillings, and leavings, and 3 per cent of spirits. This brings us face
to face with the calculation that the value of our daily waste in drinks
is nearly £6500. Carbonic acid gas is undoubtedly answerable for a
lot of this waste. In The Old Guard, a musical piece produced at the
Avenue Theatre some years ago, Mr. Arthur Roberts in his
instructions to Miss Phyllis Broughton—who made a very comely
stage barmaid—particularly enjoined her, when drawing ale, to use
her left hand to bring the handle down.
“The right hand,” he observed—of course it was all “gag”—“is for
the froth.” And then he shewed her how to make half a pint of liquor
fill a pint measure. Of course there be some professional imbibers
who would object strongly and refuse to accept the froth
programme; but on the other hand it pays the retailer, in the long-
run. I am not going to re-tell the old story of the Quaker; but will
only mention that in the early seventies the landlord of a favourite
tavern in the Strand—a house of call for histrions, which has since
then been transmogrified and adorned with much bevelled glass and
carved walnut—once confided to me that he made every bit of £300
per annum out of his froth. His barmaids were all of angelic
appearance, with most beautiful heads of hair (the girls wore plenty
of it in those days) and a wealth of pretty prattle. And the {63}
customers being susceptible, and liberal-minded, the rest was easy.
Egyptian manuscripts written at least 3000 years before the
Christian era shew conclusively that even at that primitive period the
man­u­fac­ture of an intoxicating liquor from barley or other grain was
extensively carried out in Egypt. Probably the wretched Israelites got
far more birch and bastinado than beer given them whilst engaged
in brickmaking; but it is quite on the cards that Cleopatra, when
fatigued with practising the spot stroke on her billiard-table, often
commanded one of her slaves to draw her a pint of bitter with a
head on it; and who knows but that her beloved Antony cooled his
coppers with small ale?
Pliny—who would be a useful sort of man to have in a daily
newspaper office nowadays—records that in his time a fermented
drink made from “corn and water” was in regular use in all the
districts of Europe with which he was acquainted. But in Britain little
was known about beer before the Roman conquest, as the favourite
beverages of our ancestors were mead and cider. But the Romans,
although they never quite succeeded in subduing the stubborn
dispositions of the “barbarians,” managed to teach them a bit of
husbandry, and to shew them something about brewing. There were
no means of making wine in those days, and—save in Wales—there
were no grapes to make it with; but the Latins were not long in
teaching the Britons—who were never slow to learn anything which
might lead to revelry—that a very good substitute for wine might {64}
be expressed from grain and water. Hops were undoubtedly known
in England before the conquest, but do not appear to have been
regularly used in brewing before the be­gin­ning of the sixteenth
century. It is probable, therefore, that they were employed as
medicine—and there is no better tonic than your hop. The Germans
would seem to have brewed with the “wicked weed” before the
Englanders did, according to the omniscient Pliny.
The horny-handed son of toil, who can put away his four or five
gallons daily during harvest-time, without falling off the waggon,
may not know it, but it is only the female hop which is used by the
brewer of to-day. The char­ac­ter­is­tics of the he-hop are not known to
the writer, or whether he plays any part in aiding to relieve the thirst
of the lieges; but the female is said to exercise “a purifying, a
preservative, and an aromatic influence over the wort.”
It used to be a popular fallacy that the beer made at Burton-on-
Trent was brewed from Trent water, instead of, as was and is the
case, from spring-water, which is eminently suited to the purpose.
The chief industry at Burton was, originally, cotton-spinning, but fifty
years ago this industry was discontinued owing to the triumphal
march of John Barleycorn. Why spin cotton when the man­u­fac­ture of
beer is not only a much healthier occupation but is far more
lucrative? So Burton stuck to its beer-making, a trade which was
originally established there—in a very small way—in the sixteenth {65}
century. There appears to have been a demand for Burton ale in
London, during the reign of Charles I.; although details are missing
as to whether the demand extended to the royal palaces. It is
certain, however, that more than one hundred years ago Burton-on-
Trent did a considerable export trade with the Baltic. In 1791 there
were nine breweries here, and in 1851 sixteen. But at the be­gin­ning
of the present century, until the last-named year, when the great
Exhibition attracted all the world and his wife to England, the
breweries at Burton were not all in a flourishing condition; and I
have more than once heard my grandfather—who spoke from
personal knowledge—tell the story of how the late Mr. Michael
Thomas Bass most magnanimously offered to “prop up” another
large firm, with the remark, “There’s room enough for us both here!”
 At present there are thirty breweries in Burton-on-Trent, and
employed in these are some 8000 men and boys. After the opening
of the Midland Railway in 1839 the brewing trade here began to
improve, but it was mainly due to the energy and practical
knowledge of Mr.
Bass
aforementioned that Burton-on-Trent in general, and the great firm
of Bass are in their present flourishing condition. In the words of
Shakespeare, “He was a man; take him for all in all we shall not look
upon his like again.” Beginning as traveller to the firm, he was not
long ere he became its chief director. He was untiring in {66}
business, a man possessing the broadest views of men and things, a
bit crotchety on occasion, but possessed of “that most excellent gift
of charity,” in boundless supplies. Amongst his other benefactions
was the building and endowment of St. Paul’s, Burton, and the gift
of recreation grounds, a free library, and swimming-baths to the
adjacent town of Derby. He also built and endowed another church
on his own estate, at Rangemore; and his hand was never out of his
cash-pocket when he could aid in a good work. He represented
Derby, in the Liberal interest, from 1848 to 1883, and was a tower of
strength to that party, albeit possessed of nothing like bigoted
opinions. On the contrary, it was his custom through life, like Hal o’
the Wynd in The Fair Maid of Perth, to “fight for his own hand.” And
as an instance of his energy and grit, it may be mentioned that after
voting in the House of Commons for Mr. Gladstone’s Irish Church
Disestablishment Bill—the division on which did not take place till 2
A.M. —he travelled by the “newspaper train” at 5 A.M. from Euston to
Rugeley in order to hunt with Mr. Hugo Meynell Ingram’s hounds,
later in the morning, changing his clothes on the way down. The
meet was at Brereton Hayes, close to Cunnock Chase, and I well
remember greeting him that morning, and receiving for a reply:
“Thank you, I’m pretty well for an old ’un.” He was over seventy (I
think) at the time. That was three decades ago; and since then the
trade of Bass has increased enormously. For the annual holiday {67}
of the staff I should be afraid to state from memory how many
special trains are required to convey the great hive of workers to
Brighton, and other far-distant watering-places, and back to Burton
again. In short, it would be hard to find a spot in the inhabited world
in which the name of Bass is not known and respected.
I mentioned further back Scotch and English barleys as being
employed for malting purposes; but as a matter of fact the produce
of many countries is used, in a blend, the whole being divided into
two classes, heavy and light. And in making choice of barleys it is
necessary that they should be thoroughly and equally ripened, well
“got” or harvested, and as far as possible presented to the brewer in
the perfect husk or envelope with which nature has furnished the
kernel. Ancient and modern modes of thrashing and dressing to a
greater or less extent damage both the husk and the kernel, and
thus at the very threshold introduce one of the causes of disease.
Whenever the grain is broken or bruised it is liable to be attacked
when moist by a variety of moulds which lead to more or less
serious disaster.
Of the different varieties of beer, “pale ale” or “bitter” is a highly-
hopped beer made from the very finest selected malt and hops;
whilst “mild ale,” or as it is called in Scotland “sweet ale,” is of
greater gravity or strength, and is comparatively lightly hopped. “Old
ale” is, naturally, the best stuff that can be brewed, in a state of
maturity; and it is a peculiarity of ale that, securely bottled, it will {68}
keep its strength far longer than any other fermented drink. In
December 1889 some bottles of beer were found walled up in a
cellar at Burton-on-Trent; and the records of the firm, as well as the
shape of the bottles, shewed that the beer had been brewed nearly
a hundred years before. It was as bright as a sunbeam, and quite
drinkable, but had lost its bitterness, and assumed the character of
sherry. But old ale, like old brandy, is of little value to the toper, in
that it takes a very minute quantity to accomplish in him the desired
effect—oblivion. “Audit” ales and “college” ditto require very delicate
handling of the jug; and I have tasted ancient beer in Allsopp’s
cellars in Burton, a wine-glassful of which would probably have put a
coal-whipper on his back. It was the colour of mahogany and oh! so
seductive.
Porter, as most people know, is a black beer, brewed in much the
same manner as the other stuff, with roasted malt to give it colour;
whilst stout is simply a superior kind of porter. As for the lager beer
of the Fatherland it is fermented at a very low temperature, the
fermentation being longer delayed. Some years ago great stress was
laid on the German system of mashing called the “thick mash,”
which consisted of boiling or cooking a portion of the mash, and
running it back and remixing it with the portion left in the tun; but it
is now found possible to brew the finest lager beer with a slight
modification of our own mashing method.
The sons of Britannia for a considerable period held aloof from
this lager, which was pronounced by some to be mere {69}
“hogwash,” and by others to consist principally of the juice of fir-
cones and onions mixed with snow-water. The fir-cone flavour is, I
believe, accounted for by the “pitching” of the barrels in which the
beer is stored; but I don’t know where the oniony flavour comes
from. The prejudice against this beer has long since departed from
our midst; in fact it has become quite a favourite summer drink. It is
generally considered less intoxicating than its English cousin. In fact
the German students are in the habit of putting huge quantities
thereof out of sight, on the occasion of passing examinations, and
public rejoicings; and these “beer-drinkings” are, apparently, fully
sanctioned by the authorities.
It has been written that it is to Dean Nowell, “classed by Fuller
among the worthies of England,” that we are indebted for the
discovery of bottled beer. According to Fuller, “this worthy, who was
an en­thus­ias­tic fish­er­man, was one day angling in the Thames; but
at the very time when he was trying to catch perch to carry to the
frying-pan, that benighted bigot Bishop Bonner was trying to catch
him to tie him to the stake for pur­poses of cremation, to the glory of
the old religion. The reverend gen­tle­man heard that he was ‘wanted,’
left his fishing, and fled as far from the Thames as he could, leaving
untasted in a safe place a bottle of beer which he had filled in the
morning. Bonner’s day did not last long, and Dean Nowell was soon
able to return to his old haunts. Fishing as usual, he went to look
after his bottle of beer, and found that it had turned into a {70}
species of gun—it exploded its contents, when touched.” Thus
Nature, which is ever kind, turned the martyrdom and misery of
Bloody Mary’s reign to good—it brought about bottled beer. The
Dean un­bos­omed himself of his great dis­covery to his clerical
friends, and the clergy let it out gradually to the laity.
 Gervase Markham, the aforementioned con­tem­po­rary of Shake­‐
speare, gives the fol­low­ing directions to “the English House­wife” of
1631, for
Brewing of Bottle-Ale.
Touching the brewing of Bottle-ale, it differeth nothing at all from the brewing
of strong Ale, onely it must be drawne in a larger proportion, as at least twenty
gallons of halfe a quarter; and when it comes to be changed, you shall blinke it
(as was before shewed) more by much than was the strong Ale, for it must bee
pretty and sharpe, which giveth the life and quicknesse to the Ale: and when you
tunne it, you shall put it into round bottles with narrow mouthes, and then
stopping them close with corke, set them in a cold sellar up to the wast in sand,
and be sure that the corkes be fast tied in with strong packe-thrid, for feare of
rising out, or taking vent, which is the utter spoyle of the Ale.
Now for the small drinke arising from this Bottle-ale, or any other beere or ale
whatsoever, if you keep it after it is blinckt and boyled in a close vessell, and then
put it to barme every morning as you have occasion to use it, the drinke will
drinke a great deale the fresher, and be much more lively in taste. {71}

I confess that the above directions are somewhat vague to my

untutored mind, which is quite a blank upon the subject of “blinckt
and boyled” ale. Nor do I imagine for one moment that the “English
Housewife” of the year 1899 will cumber herself with brewing or
bottling, any sort of malt-liquor, as long as there be bonnets to be
chosen, bicycles to be ridden, or golf to be played.
Wholesome as may be the beer in itself, its surroundings are not
always hygienic. The system of pumping up the glorious fluid from
the cellar through leaden pipes neither improves the flavour nor
renders it more valuable as a morning “livener.” And there is a story
—which I believe to be strictly true—told of a night cabman in
London who used to call at the nearest tavern to his stand, the first
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!

ebookultra.com