Customer Reviews

A Course in Digital Signal Processing

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

 

 

I am using the text in a DSP course as an undergrad in Electrical Engineering. It is a very clearly written, precise text that I feel is written for the undergraduate/first-year graduate who is taking a first course in DSP. Because of this, the author tends to omit some important details that other books include, particularly Oppenheim and Schafer (the intro DSP bible, as it were). This can be a great help to busy students who want a basic understanding of the theory behind DSP without caring too much about its nuances.

The disadvantage of the author's conciseness is a general lack of thorough descriptions of complicated ideas. Many ideas are paraphrased rather than explained in full, which I suppose is natural at this level. For example, when discussing the concept of sampling random signals, Porat notes that "White noise cannot be sampled" because it "does not exist as a physical entity." This statement, while true, is incomplete, as any knowledgable DSP expert can attest.

That MATLAB code, while present, is sparse and incomplete. I suppose this is better than nothing, but don't expect a computer-based approach to problems!

Otherwise, this book has proven a solid, easily understandable (with the proper math background) textbook for a first-level course in DSP after a standard Signal Processing course. If you're looking for more detail, go with Oppenheim and Schafer.

While the discussion in this book is probably all accurate, the book has two problems that make it better suited as a supplemental reference than a course text. First, the author uses a very strange notation for transforms that seems to be of his own devising. Second, a number of the problems are poorly written making it very hard to see what he is even asking. A third problem worth noting is that there doesn't seem to be a Fourier transform table anywhere in the book. If you are coming to this book from other DSP texts or Signals and Systems books get ready to work just to figure out what he's talking about.

I liked this book. The author starts with a review of frequency domain analysis (good in case it has been some time you dealt with that, like me) then moves on in consecutive chapters with a reasonable amount of examples for every topic. I liked the clear presentation of the Z-transform. I also use the book now as a reference book although it's not optimal for that purpose. The author supplies many exercise problems at the end of each chapter, but most of them are quite hard to solve, especially if you have no other background than the book itself.

Several things I liked about this book:
a. Concise.
b. The order of the book (better than oppenheim's book). CTFT, DTFT, Sampling Theorem, DTF, FFT, Spetral Analysis are all related to Fourier analysis. Oppenheim's book has filter desgin in between, I don't know why he did that.
c. I like the examples and questions provided in this book. For example, the example about the Nyquist-T spectrum, the example about the passband signal direct sampling...very thoughtful, useful examples for engineers.
Things I don't like:
a. The book is kind of weak on Sampling theorem and DFT. I don't know why the author wants to cut DFS, i think it's quiet useful to understand DFT.
I don't recommend this book for self-study (oppenheim's book is better with more details). However, it's a good concise textbook given the condition that there is an experienced intructor guiding the students. I'd like to use this book as textbook if I become a professor in the future.

The author has managed to cram an impressive range of topics into a relatively slim volume. He does so, of course, by covering hardly any of them in any appreciable depth. The textual descriptions are, as the other reviews point out, generally confusing and of little value; the meat of the book is in the proofs that accompany each result. Unfortunately, these too sometimes use an unconventional notation and worse, are poorly typeset: there are often two or three steps combined into a single line and it will take the reader a few passes to figure out what is being shown.

Once you've worked through the proofs, you're hardly done; many of the most interesting corollaries of each main result are discussed only in the problems following that topic. The problems are tersely and often vaguely stated, and there are no hints or answers given anywhere in the text. Many have an open-ended structure (i.e. asking "What can be inferred from ..." instead of "Show that ..."), which is a great format for an interactive seminar but absolutely horrible for self-study. Many of the problems can be solved only after having had the proper insight on a previous problem, but there is no explicit clue to this dependence in most of the problem statements.

The target audience of the book is not clear; the lack of guidance, clear explanation, or useful examples make it clearly inappropriate as an introductory text, but the terseness and the fact that so many important results are buried in unanswered questions makes it of questionable value as a reference for those already well-versed in the field. Some of the problems might provide interesting ideas for lecture topics or exam questions, but the book itself has little value as a study aid. Highly recommended for sadistic professors trying to carve out a reputation as the toughest SOB in the department or math enthusiasts looking for a way to kill some time during a long prison sentence; everyone else should steer clear.

I LOVE this book for its exemplary clarity, conciseness and rigor. I liked having both the informal proofs which involve distributions given in parallel with the elementary proofs, thus presenting discrete signals both as sequences and as distributions. The book is written without any annoying mathematical sloppiness, quite common in DSP literature; yet, it is entirely accessible for everyone who is willing to invest an appropriate effort. Also, this is a real TEXTBOOK rather than a reference book, which keeps its size manageable and non-intimidating. In short, in my opinion, this is a must-have book for any student wishing to learn DSP with genuine understanding.

I'm an electronic engineer.
I wanted to learn how to design digital filters. This book is very straight forward and all the matlab macros are included so that you can implement the filter design functions by yourself. I used visual basic in Excel. Its the best book so far I have read on how to design digital filters and what is the difference between the kind of filters and windows you choose. Now it is much easier to understand how to use the Matlab functions needed. It thorougly describes pros and cons of different filter types. Both FIR and IIR filters are covered.
For IIR filters different implementations are described.
A little about number of bits needed for coefficients.

Using a systematic approach this book starts at the very basics of signal processing through showing the analogies between window and filter designs and ending at more advanced topics such as quantization effects and random signal processing.

While the notation used in this book is not considered standard, it is consistent and saves from errors (the d for the DFT transform clears many ambiguities of other notation systems)

This book is best fit for advanced undergraduates and graduate students as well as anyone dealing with signal processing at a theoretical level. I have used this book as a textbook on my first course in DSP and later based upon its knowledge wrote a lab experiment in DSP and in particular consequences of quantization effects.

This is a good overall book although the notations while consistent were different from most other books. The key part of this book are its problems. The explanations of concepts are sometimes only fully understood after reading through the problems, understanding them and solving them. The difficulty level of the problems varies but for the most part reflect the text in the chapters. The asterisk annotated problems are harder but have a lot of practical significance. The organization and sequence is definitely better than Oppenheim Schafer. Covering Sampling Theorem, DFT,FFT first is definitely easier and provides a solid base to start understanding Z transforms and filter design. So if you are a immune to the notational differences or are ready to take on different notations (some people are not), this is a good book for you.

I took DSP basic course at my final B.Sc.EE year intending to take DSP as one of my majors. The book proved to be an excellent study resource (IMHO, of course), providing the main DSP concepts in very clear manner backed up by an examples. The stuff is presented in very thoughtfull sequence, similar to regular DSP basic course sillabus, explanations are easy to comprehend and math is easily understood (bearing appropriate math background).
I used the book solely for the study of the course, found it to be the best learning core, allows you to grasp DSP fundamentials, at least to the level necessary for basic DSP B.Sc.EE course.
The book material stratches far beyond the basic DSP course that is usually taught for B.Sc.EE, however I have yet proceeded to that material, hence cannot vouch for that material, albeit intend to go further into it later.
Hadn't chance to compare this book with its main rival (Oppenhaim and Shafer), however so far found Porat's book leaving very little to be desired to look over to others.
Bottom line - highly recommended for B.Sc.EE level DSP study and probably beyond that.