An asymmetric neural network successive approximation A/D converter

Summary form only given, as follows. Analog-to-digital (A/D) converters have enabled various analog signals to be processed by today´s flexible, programmable, fast, and accurate digital devices such as digital computers. There are several existing designs of digital A/D converters, such as the successive-approximation A/D converter and the symmetric neural network A/D converter which was introduced by Hopfield and Tank. A novel A/D converter using an asymmetric neural network, which shows simple structure and advantages in conversion speed and accuracy over the other A/D converters, has been developed.<>