A decentralized neural architecture based A/D converter with binary coded outputs

Abstract

In this paper, a new approach is given to design a 4-bit decentralized flash analog to digital converter (ADC) with binary coded outputs. Four artificial neural networks (ANN) are chosen and trained to identify an ideal 4-bit ADC. In order to test the obtained neural ADC, simulation results are carried out to show the efficiency regarding to parallel encoders and neural ADCs already proposed.