Measuring the complexity of metrological systems' signals

This paper outlines the possibility of attaching the dimensions of complexity to signals in metrological systems. Deterministic chaos processes can produce signals with broad-band spectra similar to random processes. In the former case, chaotic systems have finite dimension of complexity, in the latter case true random processes ought to be infinite-dimensional. Given observed time series, a finite real number for calculated fractal dimension has been assumed to imply that a deterministic complex processes produced the observed time series. Signals generated by precision voltage sources, in our case produced by a solid state voltage reference elements (VRE-s) of a group DC voltage reference module (DC VRM) are treated like complex system's signals. DC VRM consists of a parallel group with four VRE-s and a serial group of two elements acting as a master reference MR. By calculating the fractal dimensions, we are trying to understand and to classify the behaviour of each VRE and the behaviour of the whole group reference module DC VRM as a complex system. In order to understand the physical interpretation of the results of the fractal analysis and relate it to the common metrological requirements, the results are compared with typical statistical parameters.<
>