Digital Signal Processing Reference
In-Depth Information
Chapter 8
Conclusion
Perhaps the best place to end is where we started. In the beginning, we set out to
prepare and present a collection of predesigned filters for development
applications specifically for users who cannot find the time to explore the wide
range of filter design techniques available. It is our belief that time to market is
critical in product development and as such predesigned filters present
themselves, perhaps not the best, but as an immediate solution. Given the wide
range of filter applications and specialized solutions such an approach is
ambitious and therefore remains a practical limitation.
In order to maintain a manageable collection of filters we restricted ourselves
to the simplest technique, the window method. The Gaussian window was chosen
for its simplicity and resonance with scientists, engineers, statisticians,
economists, and other product development practitioners. Hardly any theory was
presented to back up any claims made, since there are several excellent texts on
digital filters and signal processing. However, it was thought that a book of this
nature should cover some basic filtering tools with suitable examples for a smooth
transition into the approach. This was dreamt up to keep certain basic tools of the
trade at hand and in one place. In this context, therefore, approximately 260 filters
were characterized and presented both graphically and in tables. Some more
contextual details were given for differentiators and Hilbert transformers.
It is hoped that our approach brings many financial and other rewards to the
scientific community at large, and a significant improvement to product
development throughput. This could also be an enjoyable, practical learning
experience for many first-time users, as well as students who are keen to apply
digital filters. Time will tell.
457