Customer Reviews

Lessons in Estimation Theory for Signal Processing, Communications, and Control (2nd Edition)

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

 

 

I used this book for a class during my PhD in Elec Engr. This book is best read with a second book in statistics and a third on signals close at hand. The concepts are complex and presented in a reasonably brief manner. It packs a lot of punch in a relatively small space. If graduate classes in signal processing and control systems are your bag, this may be the book for you.

I first used this text book in 1980 at USC. At that time, it was only in the form of prepared lecture notes. The current organization of this book is the same as before. New material has been added. I purchased this book because of the new revised material. I find the book to be very systematic. It is an excellent book for personal study of the subject or as a graduate text book. Since 1980, there are some other books that are now available on the subject. But this book stays very relevant and useful as ever. I recommend it highly.

I am a student of the author and I really enjoyed this book along with his lectures. I may not enjoy reading this book without going through the lectures of the author (This is the scenario with many subjects to most students). The book is well written well organized and I benefited a lot from this book.

During the winter 2008 academic quarter I took an introductory class on optimal estimation which relied heavily on this book, covering about 70% of the material. I felt the text gave a great background on optimal estimation, but because I used it in conjunction with an engineering class, I can't really say how great the book would be as, say, a self-study aid. I am disappointed at the price vs. binding ratio for this book, since at $99 I would expect a book like this with no glossy color pictures to be hardcover, whereas it is soft cover. The content of the book itself is great.

The book is divided into many very short chapters, which I really appreciate, since it builds up in small, manageable increments the foundation to more complicated optimal estimation concepts. For example, one chapter will cover Gauss-Markov processes and their properties, while the next will cover Kalman filtering. The homework problems, which aren't too difficult, have a few typos, but it was obvious they were typos. All in all, for me this book provided a clear introduction to optimal estimation, but your mileage may vary if you are using it for self-study.

I don't know after I finished this book. And reading this book is definitely not an enjoyable experience. The worst thing is the writer wrote this book into a math book. The worst of the worst, it's a math book with 'n', 'N', 'k', 'lambda'... all these messed up. It takes unnecessary effort to figure out the dimensions of matrixes and vectors. After all the torture, I cannot find a tiny bit of physical meaning attached behind. The problems are remotely connected with the text. I cannot do the problems after I read the text twice, or I cannot even understand what the problems are talking about. As an engineering student, I believe the underlying essence of any theory is simple. It's the writer used 560 pages to make it complicated and confusing and the student now have to do the reverse work.