A full neural Gray-code-based ADC

In this paper a new method of designing Gray-code-based analog-to-digital converters (ADC's) is presented. The method, derived from research on the neural networks, has the aim of replacing all functional components needed in the ADC block scheme by a simple connection of analog neurons. First, the theory underlying the method and a functional description of the new neural ADC are given. Then, in order to highlight some of the design considerations, a simulation study of the ADC is provided. Finally, an 8-bit discrete-component prototype is discussed, and the experimental results are also given.