Practical Applications in Digital Signal Processing

Richard Newbold’s Practical Applications in Digital Signal Processing contains practical solutions for common DSP applications – solutions that have been passed along by word of mouth and/or technical papers, but have somehow eluded textbook authors… until now.

It’s written for upper-level or graduate students in electrical and electronic engineering. Practicing engineers with a background in DSP will find solutions to common DSP design problems. This is NOT a beginner’s book; readers should already have at least a basic knowledge of digital signal processing.

Review of Basic DSP and Mathematical Concepts

Practical Applications in Digital Signal Processing begins with a review of basic DSP concepts such as frequency and sampling of sinusoidal waveforms. Clear diagrams accompany equations and the narrative, as the author describes the quantification and digitization of a waveform from both a theoretical and practical perspective. Newbold discusses sampling rates, the Nyquist rule, and aliasing. The review is brief, so if those terms aren’t already familiar to you, then you should find an introductory DSP book to get you started.

The next few chapters include a review of mathematical concepts related to DSP such as complex numbers, the Fourier Transform, and the Z-Transform. The author states that it’s not his intention to duplicate the “excellent material” of other authors. Instead, he directs the reader to a thorough list of sources at the end of each chapter. That’s fine, but I wonder if it would be better to simply refer to that list in lieu of those review chapters, which represent over 100 pages of math with almost no mention of DSP. On the other hand, professors using this book might want to have the students read the book out of sequence, beginning with the applications and referring to the background review material on a just-in-time basis.

FIR Filtering Techniques

The book shines with the chapter on Finite Impulse Response (FIR) digital filtering, one of the most common applications of DSP. Newbold discusses filter types, their applications, and attributes, and provides a nice summary table (below).

He then introduces the Parks-McClellan (PM) algorithm, including the complete subroutine written in its original FORTRAN. While modern DSP applications are programmed in C or C++, the author suggests using the tried-and-true PM algorithm in FORTRAN, and he gives instructions for embedding that code into a C/C++ program. An appendix gives a listing of the complete PM algorithm (reprinted with permission) – all 25 pages! It would be nice to have a downloadable version available online. Instead of transcribing the entire subroutine, you might consider scanning it with optical character recognition (OCR) software. You’ll get the satisfaction of using DSP technology (part of OCR) to obtain a DSP algorithm!

In addition to the mathematical analysis of each filter, Newbold provides a clear step-by-step design method and an explanation of the PM parameters that a designer must use. This is where the book really becomes valuable. One doesn’t need to understand the algorithm fully in order to design with it. The author treats it like a “black box,” giving inputs and producing outputs without requiring knowledge of exactly how the outputs are produced. Using the parameters, their descriptions, and the author’s procedure, a DSP designer should be able to design a filter using the PM algorithm.

Real-World Design Scenarios

Beginning with the chapter on filters and throughout the remainder of the text, the author presents a number of authentic design scenarios. He puts the reader into the role of design engineer, provides design specifications, works through the solution, and gives a graphical analysis of the expected output. In these design activities, Newbold writes like a skilled craftsman guiding his apprentice: casual and unpretentious, but with authority.

The Author

Richard Newbold began designing DSP-based systems when Intel was young and DSP circuits were made from discrete components and TTL chips. (For younger college students and engineers, that’s the early 1970s.) Throughout his career, he has designed circuits and algorithms for a variety of DSP applications, and compiled a sizable engineer’s notebook of DSP solutions. In this book, Newbold shares many of those solutions with his readers.

Should You Buy the Book?

If you’re an engineer with some DSP experience and you need a good handbook for designing applications such as FIR filters, numerically controlled oscillators, digital tuning circuits, elastic store memory, automatic gain control, and more, then this book should be on your desk. Also, if you’re a professor teaching an advanced DSP course, you might consider this as your textbook.

This review was written by Tom Lombardo, Professor of Engineering and Technology at Rock Valley College. Tom holds technical degrees in electronic engineering technology and computer science, as well as a doctorate in instructional technology. In addition to his teaching duties, Tom serves as Rock Valley College’s instructional designer, where he helps faculty apply new teaching methods and technologies to their classes. Tom is a member of the American Society for Engineering Education.