Evaluation of measurement uncertainty in the presence of combined random and analogue-to-digital conversion errors

摘要

The evaluation of the uncertainty in measurement it the presence of combined random errors and errors due to the analogue-to-digital conversion of measuring signals is studied. The following situation is investigated. Analogue measuring signals of a measurand under study are obtained. The signals are assumed to contain random errors with a Gaussian distribution and they are converted into the digital form. Only the resulting digital signals are to be available. The Bayesian theory of measurement uncertainty is applied to the analysis. An estimator is assigned to the measurand and the distribution of this estimator is determined. This distribution can then be used to numerically calculate a best estimate of the measurand and its associated uncertainty. The application of this approach to some examples is presented and compared with results obtained by treating random and analogue-to-digital conversion errors additively and independently, which is frequently done. In some cases, remarkable differences are found which can justify the increased time and effort involved in the approach proposed.