Synthesis and implementation of T-model neural-based A/D converter

Describes the design, implementation, and experimental results for a T-model neural-based A/D (analog-to-digital) converter. The converter architecture is presented with particular emphasis on the elimination of local minima of the Hopfield neural network. It is well suited to serve as a solver of a variety of optimization problems. The converter is synthesized with a simple learning algorithm and fabricated in standard CMOS circuits. Experimental A/D converter results are presented; these results verify the full functionality and their operations.<>