An 18-bit analog-to-digital converter has been designed that uses the principle of oversampling rather than the successive-approximation technique of most converters. The design employs a two-stage digital filter/decimator to reduce the sampling rate. Unusual digital filter structures are used to reduce the computation rate to a practical level. Extension to 19 bits and beyond is discussed.
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.