It's a bit surprising that after 202 upvotes and 51 comments nobody has written any comments about the contents of the linked book. Well, I guess I have to go out on a limb and say it:
This book is embarrassingly awful.
> The Fourier transform is just a different way to describe a signal. For example,
investigating the Gibbs phenomenon, which states if you add sine waves at specific
frequency/phase/amplitude combinations you can approximate a square wave, can
be expressed mathematically as (2.1),...
1. The Gibbs phenomenon doesn't state anything. It can be observed; it exists; it is known; it entails things; it implies things; you can state things about it; but it does not, itself, state anything. It is as mute and voiceless as a nudibranch.
2. What is being described is not the Gibbs phenomenon, but the Fourier transform of a square wave. Now, it is true that the truncated Fourier transform of a square wave does exhibit the Gibbs phenomenon. But so does the truncated Fourier transform of any other function containing discontinuities. It's not specific to square waves.
3. This paragraph starts off trying to ineptly explain what the Fourier transform is (a different way to describe a signal? Different from what?), despite the fact that the previous paragraph provides a different and better explanation which is still not very good, but somehow forgets it was going to do so before getting sidetracked with committing the appalling category errors above on the Gibbs phenomenon.
4. You might reasonably assume that the above paragraph, even if poorly expressed, would be followed by some kind of discussion of what the Gibbs phenomenon is, when it matters, and/or when it arises. You would be disappointed; evidently the authors couldn't remember their undergraduate Signals & Systems courses well enough to remember what the Gibbs phenomenon was.
I had skipped ahead to chapter 2 because I couldn't deal with the mindnumbingly awful crap they stuffed Chapter 1 full of, but it didn't get any better. I have to assume the whole book is this literarily abominable and careless about accuracy.
Analog Devices' apparent choice to promote this book for training material suggests that the company's quality standards are no longer what they historically have been and what allows them to charge such a premium for their parts.
Is there some part of this book that is of passable quality? Please tell me that after some introductory filler chapters they started trying to write coherent sentences? I don't have the heart to check.
To all appearances, the HN comments on this post are both better written and far higher in information density than the book it's ostensibly about.
Alex was my advisor in grad school and I only know him to be honest, hard working and technically sound. I haven’t read this book closely but I’m not convinced that it’s poorly written based on cherry picking one paragraph from it.
Maybe someone else was sabotaging his work. Is it possible to cherry-pick one paragraph from it that is well-written and technically sound? Because I gave up after Chapter 1 and the beginning of Chapter 2. Maybe you could ask Alex what he thinks of how the book turned out and post his answer here?
I agree that it's mean, and I feel kind of bad about it, but I think the interests of the hundreds of people who are apparently interested in reading a book about software-defined radio outweigh the interests of the four authors to not have their feelings hurt. The problem is not that they've written this colossal pile of garbage; there's nothing wrong with that, and it's often the first step toward writing something worth reading. The problem is that that pile is being promoted (by Analog Devices, no less) as a useful way to learn about software-defined radio.
This book is embarrassingly awful.
> The Fourier transform is just a different way to describe a signal. For example, investigating the Gibbs phenomenon, which states if you add sine waves at specific frequency/phase/amplitude combinations you can approximate a square wave, can be expressed mathematically as (2.1),...
1. The Gibbs phenomenon doesn't state anything. It can be observed; it exists; it is known; it entails things; it implies things; you can state things about it; but it does not, itself, state anything. It is as mute and voiceless as a nudibranch.
2. What is being described is not the Gibbs phenomenon, but the Fourier transform of a square wave. Now, it is true that the truncated Fourier transform of a square wave does exhibit the Gibbs phenomenon. But so does the truncated Fourier transform of any other function containing discontinuities. It's not specific to square waves.
3. This paragraph starts off trying to ineptly explain what the Fourier transform is (a different way to describe a signal? Different from what?), despite the fact that the previous paragraph provides a different and better explanation which is still not very good, but somehow forgets it was going to do so before getting sidetracked with committing the appalling category errors above on the Gibbs phenomenon.
4. You might reasonably assume that the above paragraph, even if poorly expressed, would be followed by some kind of discussion of what the Gibbs phenomenon is, when it matters, and/or when it arises. You would be disappointed; evidently the authors couldn't remember their undergraduate Signals & Systems courses well enough to remember what the Gibbs phenomenon was.
I had skipped ahead to chapter 2 because I couldn't deal with the mindnumbingly awful crap they stuffed Chapter 1 full of, but it didn't get any better. I have to assume the whole book is this literarily abominable and careless about accuracy.
Analog Devices' apparent choice to promote this book for training material suggests that the company's quality standards are no longer what they historically have been and what allows them to charge such a premium for their parts.
Is there some part of this book that is of passable quality? Please tell me that after some introductory filler chapters they started trying to write coherent sentences? I don't have the heart to check.
To all appearances, the HN comments on this post are both better written and far higher in information density than the book it's ostensibly about.