Classification and digital synthesis using spectral techniques
Date
1984
Authors
Iyer, Balasubramanian H.
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This thesis is concerned with the effectiveness of digital circuit design and fault detection, and concentrates on the investigation of several techniques for the classification and synthesis of Boolean functions. A more compact representation of individual Boolean functions based on spectral classification using only O(n2 ) coefficients is derived.
Digital design based _on spectral classification has the benefit that all functions can be realized using a small set of prototype functions. The resulting modular design is based on the implementation of the prototype function together with the appropriate transformations required to map the given function onto the prototype function. A new design philosophy utilizing symmetry techniques to implement the prototype function supplemented by appropriate spectral translations is developed. Its limitations for both synthesis and fault detection are investigated.