IMPROVED CONVERGENCE IN A DIGITAL SIGNAL AVERAGER

摘要

A method for improving the speed of convergence to sub-least-significant-bitaccuracy and precision in a digital signal averager when the measurements areaveraged over many samples where the noise inherent in the signal is not largecompared to 1 LSB. A slider offset circuit (10) implements the method whichadds offsets to an analog input signal acquired from an analog sensor (20).The slider offset circuit (10) compensates for limitations in the resolutionof an analog-to-digital converter (28) by adding a sub-least-significant-bitoffset to the input signal. Similarly, the slider offset circuit (10)compensates for the differential non-linearity using a Gatti offset which arewhole number multiples of the average voltage bin width. Speed of convergenceto sub-LSB accuracy is further improved by employing non-consecutiveincrements in the two sliders (22, 24).