An algorithm for symbolic and numeric architecture determination in a knowledge-based analog-to-digital converter synthesis environment using fuzzy membership functions

This paper introduces fuzzy membership functions as a solution to architecture determination problem in a knowledge-based system level synthesis process of analog-to-digital converters. The knowledge-based synthesizer is to determine the architecture to be used based on a set of symbolic or numeric specifications provided by the designer, examples are low power consumption, high speed and low resolution. The other set of possible specifications is that of numeric values, when the designer has information about the speed, power or resolution. Fuzzy membership functions were considered for solving that problem, because fuzzy systems have the attractive feature of being able to process symbolic as well as numeric parameters.