Customer Reviews

Fundamentals of Adaptive Filtering

5 star:		(8)
4 star:		(1)
3 star:		(0)
2 star:		(1)
1 star:		(0)

 

 

This is the best textbook I've ever read. The mix of theory and practice, examples and math, and the ordered introduction of concepts flows so well it reads like a treasure quest. Everything has its place, and the place is known - you just have to follow the signs and keep an eye out for those which are subtler than others, just to keep things alive and interesting. It's indicative of the clarity of thinking and broad sweep of this work that you would refer to it (I do and other students I know as well), for other subject matter such as communications, linear algebra, digital filters, probability and control theory. The homework problems are not just a series of one-liners as in most graduate level text books, but well-marked road maps addressing points discussed in the text and opening up new possibilities. The computer projects are very well commented and the results analysed. The language is concise yet fluid, and there are extremely few typos of any kind. This is a very, very high quality work.

I work on 802.11b and 802.11a/OFDM modems and was looking for some kind of coherent text on decision feedback equalization and other tracking schemes when I stumbled upon this book. I'm deep into chapter three and I can already draw the following conclusions.

There are two things that make this book really great.
1) The teaching style. Besides having a wonderful way with words, the author uses plenty of subtle teaching principles that help in the understanding of the material like frequent repetition of the main concepts and slowly building up examples over different sections inside each chapter.

2) Content !!!! Its hard to assimilate abstract material like estimation theory without at the end having some type of useful application or algorithm to show for. The author's examples and computer projects on channel estimation, channel equalization, maximum ratio combining and Rake receivers are very popular building blocks for a variety of modems. It is very motivating to know that after labouring through the chapter the authors feels that you are now ready to understand decision feedback equalizers and so he presents you with a nice DFE computer project.

Difficult topics require competent teaching approaches and that is exactly what you will find in this book.

The only thing that would make this book even better would be a very brief introductory chapter on statistics and linear algebra. So if you are going through the book and you want to quickly revisit the definition of covariance or hermetian transpose there is one centralized section that has all the useful mathematical tools nicely listed.

As a practising engineer, I highly recommend this book.
Andreas S.

For all those of us, working in beamforming or OFDM or equalization or noise control, who find Haykin's writing uninspiring, Sayed's work has all the elegance of Prof Kailath's pedagogical style. This is one scrumptious intellectual meal. Every extant variation of the LMS and RLS algorithms is detailed, and some histroical context provided. The mathematical background needed or assumed is rudimentary. The Matlab projects are extensive in scope - yet, like most Matlab sources, they will take work in understanding their nuances.

This book is a true masterpiece. It contains the most complete treatment, theoretical and practical, of adaptive filtering that I've ever seen and I've seen just about every book on this topic on the market. In addition to the detialed theoretical derivations, this book is rich with numerous design examples and computer projects with provided matlab code. This equips the reader with proper tools to quickly understand the concepts and put them to practice with relative ease. I highly recommend this book to anyone serious about understanding adaptive filters.

I had the pleasure of taking the author's course at UCLA when the manuscript that later became this text was under development. The book is extremely well written and dedicated study of it will turn a conscientious reader into an expert in the area. One of the great aspects of the book is that not only does it offer one knowledge in the state of the art in adaptive filtering, but along the way introduces a lot of very valuable material in linear algebra, estimation, communications (equalization, etc.)...I strongly recommend this book as an alternative to Haykin or any of the other texts out there.

This is absolute model of how an engineering book should be written. Consistent, step by step despite the fact that it covers such an advanced topic. I am thoroughly impressed by the author's ability, both of clear thinking and clear writing.
[...].

This book is very mathematical and provides little to no intuition on how Adaptive filters work. Also many LMS variants are not discussed, but left as exercises and problems. For example, Leaky LMS is listed 6 times in the appendix, and 5 of those are problems.

Good for a detailed mathematical discussion, but not good for intuitive understanding.

If you have read Haykin and felt like it is the DSP equivalent of Proakis - then this book will lift your moods. The first three chapters are especially good.

Dont buy Haykin please. Its gone old and flimsy. This is a culmination of two decades of relentless addition to adaptibe filtering. I have read from original Widrow Hopf titles to Haykin bleating. but this is the culmination.

In this book, the ideas and formulas are well presented to the readers. Notations are very consistent throughout the entire book. I have enjoyed reading this book.