Introduction - Digital Filter Design Solutions - page 3

were tested in Matlab , and in situations where it was deemed significant enough,

the corresponding Matlab function is provided. Many algorithms have been

described in pseudocode ; that is, a listing of logical steps that could be easily
converted to most programming languages, such as FORTRAN and C and its
derivatives, Pascal and Basic. Thus anyone with some background in any of these
languages should be able to implement the algorithms.
Chapters 3 to 5 are different in nature than Chapter 2. These give the filter
coefficients as well as their performance features in a data sheet format. Each
filter takes up a single page in these chapters. In fact, each filter is treated as an
individual product with its associated data sheet. In this regard, the rest of the
topic should be seen as applications information. Some preliminary material is
given at the beginning of each chapter that provides information on terminology
and definitions of terms used to characterize the filters. The preliminary material
in Chapter 3 ( Low Pass Filters ) contains discussions that are largely applicable to
the other two chapters. As such, the introductory material in Chapters 4 ( High
Pass Filters ) and 5 ( Band Pass Filters ) are short, as they are unique to their
description and characterization.
Chapter 6 provides coefficients for first and second order differentiating
filters. There has been a reluctance to apply FIR differentiators because of the
noise amplification associated with such filters. In Chapter 6, we show how to
manage the noise and avoid its amplification. Again the coefficients are viewed as
individual products with specific properties. The limitations of their use are also
discussed there.
Chapter 7 ( Hilbert Transformers ) is very different from any of the other
chapters. It gives the Hilbert filter coefficients, but within the context of FM/PM
demodulation. This is a natural progression since the Hilbert transform is about
the management of phase and is best appreciated within this context. We chose
two FM/PM demodulation schemes to characterize. For those not familiar with
demodulation concepts, we recommend the use of a good text on communications
theory, although some effort has been made to be self-explanatory within the
chapter. The DC phase errors, AC phase errors, and demodulation bandwidth have
been characterized for a given set of Hilbert and low-pass filters. As such, the data
associated with a given set of filter coefficients covers two pages, with two
additional pages giving the performance of the demodulation schemes mentioned
above. In addition to this, the noise propagation through the demodulation
schemes has also been provided. We believe that given this background, the
Hilbert filter would be better appreciated and applied.

1.4 FILTER COEFFICIENTS

All digital filter coefficients given in this collection are stored and easily
retrievable from the compact disc accompanying this topic. The coefficients in
this format form the embodiment of this offering, and as such are equally