Customer Reviews

Guide to Elliptic Curve Cryptography (Springer Professional Computing)

5 star:		(3)
4 star:		(0)
3 star:		(1)
2 star:		(1)
1 star:		(0)

 

 

I bought this book because I was designing a cryptographic protocol, and wanted to know if I could use ECC in my design. It begins with an explanation of "traditional" public key cryptography (i.e., cryptography over prime fields), introduces binary fields and elliptic curves, shows how to perform computations over elliptic curves, puts this together into ECC protocols, and then includes very useful implementaiton information. This book does a good job explaining not only how to use ECC algorithms, but why they work.

As advertised, this book doesn't go into too much mathematical depth, omitting most proofs. This doesn't mean that there is no math in this book; if you don't have a decent background in algebra (no, not the stuff you learned in seventh grade), you're likely to get confused. However, if you have a little background in theoretical math and cryptography, you'll find this a very readable and easy to understand book.

The one thing that's left out of this book are intellectual property issues. Certicom owns a lot of patents on ECC, and it's not clear which ideas in this book are covered by Certicom patents. This is a minor complaint though; overall, the book is excellent. It's rare to find a book that is so exactly on target. Highly recommeneded.

This is the only source I've found that goes into the nuts and bolts of elliptic curve (EC) cryptography. The mathematical content is rich, although proofs are generally in references rather than in the text itself. The real value is in its many and detailed algorithm examples, and in the way it builds up to them.

Before it even gets into the text, Hankerson et al have created a model of clarity. In addition to the usual, front matter includes a list of abbreviations. If you've ever choked on the alphabet soup in other books, you'll appreciate how this makes the discussion much easier to absorb. There's also a list of the algorithms presented - what the practitioner wanted in the first place.

After an introductory chapter, the authors present finite field arithmetic in a thorough but readable way. First they present prime fields over the integers, then optimal extension fields and (most importantly) binary fields. There's nothing here for the cut&paste programmer, but dozens of algorithms help the thoughtful developer work through material that is immensely complicated in other presentations. Other goodies, like Karatsuba-Ofman fast multiplication appear here as well.

The third chapter is the book's real payload: EC techniques. I've been looking for years for a book that was so explicit in the how-to, without watering down the technical content. This is practical stuff - not just the theory of EC operations, but the techniques that make EC calculations practical for high-speed implementations.

The rest of the book - about half - discusses what to do with EC codes. That includes protocols for choosing parameters, public-key and signature algorithms, and standard kinds of attacks. It also includes hardware-level description of possible implementations, down to specific instruction sets and cache structures and different kinds of chip implementations. That leads to another set of discussions on attacks, the kind that go in through the power supply or RF emissions. Appendices provide or point to pragmatic details such as parameters to use or software support available.

The only thing that could be improved in this book is the index - it's just too brief, and lacks the thoroughness the rest of the book led me to expect. I hope you realize just how small a complaint that is. In all other ways, this book meets the highest expectations.

Highly recommended for anyone who needs to understand exactly how EC cryptography works, right down to the bit level.

//wiredweird

$100 for a book about implementing elliptic-curve cryptography (ECC) that doesn't discuss point counting algorithms!? The book is a hodgepodge: the first chapter talks about general crypto algorithms, the second discusses finite fields, and only the third chapter is really about elliptic-curve implementation. (The 4th chapter is about crypto protocols in general, not really specific to ECC; the last chapter is about general implementation issues but, again, not about ECC specifically.) The book would be reasonable at 1/3 the price. At this price, do yourself a favor and buy a different book.

The good:
1) touches on thing other books seems to ignore.
2) will not overwhelm you with the math

The bad/not so good:
1) the authors writing style is awful:
a) they don't care about the "logical" line of thought. they lead the reader somewhere then suddenly abandoned to wander what just happened and why.
b) they don't know when to start a new line. by cramming too much information in a paragraph sometimes combined with math forces the reader to go back and forth multiple times to make sure you didn't miss anything.

2) chapters don't stand on their own.
3) lack of figure to help convey the message in certain locations.

This book is a must have if you are interested in implementing elliptic curve cryptography. It does not have any of the juicy ellpitic curve mathematics, but that is okay as this book is directed towards engineers and others who want to learn about how elliptic curve cryptosystems are being deployed.