Converting specifications of DSP-based instruments into data suitable for Monte Carlo uncertainty estimation methods

This paper deals with the problem of convening the specifications of devices used in measurement systems based on digital signal processing into data that can be represented as probability density functions of the input quantities (samples) in a Monte Carlo procedure for uncertainty estimation. The attention is focused on data acquisition systems and input transducers. Author's proposals for facing this problem are based on the idea of considering separately the different uncertainty contributions, by defining for each of them appropriate pdf's and taking into account the correlation of their effects on the acquired samples to be processed. The advantages and limits of this approach are discussed with the help of experimental tests