Investigation of data acquisition procedures on acoustic experiments: search for reliable and abundant data

Abstract

Acoustic spectroscopy is a powerful analytical technique for the characterization of liquids. The extent of its efficacy across various applications relies heavily on the quality of the obtained data and the skillful transformation of such data into quantities of interest. However, a common practice is to conduct both acquisition strategy and signal processing based on heuristics, subjective interpretations and non clearly and explicitly justified settings and predefined parameters. This may lead to sub-optimal measurements. For these reasons, the present work investigates 3 acquisition strategies: we shed light on the averaging of acquired pulses, the synchrony of the measurement of different variables, their perceived correlation, and relative and absolute approaches to processing data. Particularly, key performance indicators are proposed while investigating the acquisition of time series for variables of interest, mainly the sample's temperature and time-of-flight. The investigated developments focus on data whose fluctuations are due mainly to temperature variations in the range of 0.01°C. It is shown that it is not necessary to keep or assume the temperature constant when proper processing of the available data is performed.