PrefaceClick here to download the full preface (Postscript file)
This book is intended as a reference for professional cryptographers, presenting the techniques and algorithms of greatest interest to the current practitioner, along with the supporting motivation and background material. It also provides a comprehensive source from which to learn cryptography, serving both students and instructors. In addition, the rigorous treatment, breadth, and extensive bibliographic material should make it an important reference for research professionals.
Our goal was to assimilate the existing cryptographic knowledge of industrial interest into one consistent, self-contained volume accessible to engineers in practice, to computer scientists and mathematicians in academia, and to motivated non-specialists with a strong desire to learn cryptography. Such a task is beyond the scope of each of the following: research papers, which by nature focus on narrow topics using very specialized (and often non-standard) terminology; survey papers, which typically address, at most, a small number of major topics at a high level; and (regretably also) most books, due to the fact that many book authors lack either practical experience or familiarity with the research literature or both. Our intent was to provide a detailed presentation of those areas of cryptography which we have found to be of greatest practical utility in our own industrial experience, while maintaining a sufficiently formal approach to be suitable both as a trustworthy reference for those whose primary interest is further research, and to provide a solid foundation for students and others first learning the subject.
Throughout each chapter, we emphasize the relationship between various aspects of cryptography. Background sections commence most chapters, providing a framework and perspective for the techniques which follow. Computer source code (e.g. C code) for algorithms has been intentionally omitted, in favor of algorithms specified in sufficient detail to allow direct implementation without consulting secondary references. We believe this style of presentation allows a better understanding of how algorithms actually work, while at the same time avoiding low-level implementation-specific constructs (which some readers will invariably be unfamiliar with) of various currently-popular programming languages.
The presentation also strongly delineates what has been established as fact (by mathematical arguments) from what is simply current conjecture. To avoid obscuring the very applied nature of the subject, rigorous proofs of correctness are in most cases omitted; however, references given in the Notes section at the end of each chapter indicate the original or recommended sources for these results. The trailing Notes sections also provide information (quite detailed in places) on various additional techniques not addressed in the main text, and provide a survey of research activities and theoretical results; references again indicate where readers may pursue particular aspects in greater depth. Needless to say, many results, and indeed some entire research areas, have been given far less attention than they warrant, or have been omitted entirely due to lack of space; we apologize in advance for such major omissions, and hope that the most significant of these are brought to our attention.
Each chapter was written to provide a self-contained treatment of one major topic. Collectively, however, the chapters have been designed and carefully integrated to be entirely complementary with respect to definitions, terminology, and notation. Furthermore, there is essentially no duplication of material across chapters; instead, appropriate cross-chapter references are provided where relevant.
While it is not intended that this book be read linearly from front to back, the material has been arranged so that doing so has some merit. Two primary goals motivated by the "handbook" nature of this project were to allow easy access to stand-alone results, and to allow results and algorithms to be easily referenced (e.g., for discussion or subsequent cross-reference). To facilitate the ease of accessing and referencing results, items have been categorized and numbered to a large extent, with the following classes of items jointly numbered consecutively in each chapter: Definitions, Examples, Facts, Notes, Remarks, Algorithms, Protocols, and Mechanisms. In more traditional treatments, Facts are usually identified as propositions, lemmas, or theorems. We use numbered Notes for additional technical points, while numbered Remarks identify non-technical (often non-rigorous) comments, observations, and opinions. Algorithms, Protocols and Mechanisms refer to techniques involving a series of steps. Examples, Notes, and Remarks generally begin with parenthetical summary titles to allow faster access, by indicating the nature of the content so that the entire item itself need not be read in order to determine this. The use of a large number of small subsections is also intended to enhance the handbook nature and accessibility to results.
Regarding the partitioning of subject areas into chapters, we have used what we call a functional organization (based on functions of interest to end-users). For example, all items related to entity authentication are addressed in one chapter. An alternative would have been what may be called an academic organization, under which perhaps, all protocols based on zero-knowledge concepts (including both a subset of entity authentication protocols and signature schemes) might be covered in one chapter. We believe that a functional organization is more convenient to the practitioner, who is more likely to be interested in options available for an entity authentication protocol (Chapter 10) or a signature scheme (Chapter 11), than to be seeking a zero-knowledge protocol with unspecified end-purpose.
At the end of the book, we have included a list of papers presented at each of the Crypto, Eurocrypt, Asiacrypt/Auscrypt and Fast Software Encryption conferences to date, as well as a list of all papers published in the Journal of Cryptology up to Volume 9. These are in addition to the References section, each entry of which is cited at least once in the body of the handbook. Almost all of these references have been verified for correctness in their exact titles, volume and page numbers, etc. Finally, an extensive Index prepared by the authors is included. The Index begins with a List of Symbols.
Our intention was not to introduce a collection of new techniques and protocols, but rather to selectively present techniques from those currently available in the public domain. Such a consolidation of the literature is necessary from time to time. The fact that many good books in this field include essentially no more than what is covered here in Chapters 7, 8 and 11 (indeed, these might serve as an introductory course along with Chapter 1) illustrates that the field has grown tremendously in the past 15 years. The mathematical foundation presented in Chapters 2 and 3 is hard to find in one volume, and missing from most cryptography texts. The material in Chapter 4 on generation of public-key parameters, and in Chapter 14 on efficient implementations, while well-known to a small body of specialists and available in the scattered literature, has previously not been available in general texts. The material in Chapters 5 and 6 on pseudorandom number generation and stream ciphers is also often absent (many texts focus entirely on block ciphers), or approached only from a theoretical viewpoint. Hash functions (Chapter 9) and identification protocols (Chapter 10) have only recently been studied in depth as specialized topics on their own, and along with Chapter 12 on key establishment protocols, it is hard to find consolidated treatments of these now-mainstream topics. Key management techniques as presented in Chapter 13 have traditionally not been given much attention by cryptographers, but are of great importance in practice. A focused treatment of cryptographic patents and a concise summary of cryptographic standards, as presented in Chapter 15, are also long overdue.
In most cases (with some historical exceptions), where algorithms are known to be insecure, we have chosen to leave out specification of their details, because most such techniques are of little practical interest. Essentially all of the algorithms included have been verified for correctness by independent implementation, confirming the test vectors specified.
This project would not have been possible without the tremendous efforts put forth by our peers who have taken the time to read endless drafts and provide us with technical corrections, constructive feedback, and countless suggestions. In particular, the advice of our Advisory Editors has been invaluable, and it is impossible to attribute individual credit for their many suggestions throughout this book. Among our Advisory Editors, we would particularly like to thank:
In addition, we gratefully acknowledge the exceptionally large number of additional individuals who have helped improve the quality of this volume, by providing highly appreciated feedback and guidance on various matters. These individuals include:
Peter de Rooij
S. Mike Matyas
We apologize to those whose names have inadvertently escaped this list. Special thanks are due to Carrie Grant, Darrel Hankerson, Judy Koeller, Charles Lam, and Andrea Vanstone. Their hard work contributed greatly to the quality of this book, and it was truly a pleasure working with them. Thanks also to the folks at CRC Press, including Tia Atchison, Gary Bennett, Susie Carlisle, Nora Konopka, Mary Kugler, Amy Morrell, Tim Pletscher, Bob Stern, and Wayne Yuhasz. The second author would like to thank his colleagues past and present at Nortel Secure Networks (Bell-Northern Research), many of whom are mentioned above, for their contributions on this project, and in particular Brian O'Higgins for his encouragement and support; all views expressed, however, are entirely that of the author. The third author would also like to acknowledge the support of the Natural Sciences and Engineering Research Council.
Any errors that remain are, of course, entirely our own. We would be grateful if readers who spot errors, missing references or credits, or incorrectly attributed results would contact us with details. It is our hope that this volume facilitates further advancement of the field, and that we have helped play a small part in this.
Alfred J. Menezes
Paul C. van Oorschot
Scott A. Vanstone
Return to the HAC home page