Ultrasonic data acquisition: sampling frequency versus bandwidth

Abstract

Digital ultrasonic systems offer variety of advantages over conventional systems. Signal conversion into digital for allows for easy and flexible signal processing organisation. Choice of sampling frequency usually is the first question arising during such systems design. When bandwidth of the system is already known, sampling frequency can be easily determined using Nyquist sampling criteria. But usually it is desired to have lower sampling rate than necessary. Sampling rate reduction allows for computation time increase and computer memory requirement decrement. It must be noted that in such case sufficient filtering has to be applied for circular convolution results to be taken on account. Despite A/D conversion sampling rate can be chosen slightly lower than required limit, care must be taken for possible artifacts. Here, we have proven, that -6dB /-3dB bandwidth calculation is not enough for optimal choice of sampling frequency. The signal frequency characteristics have to be measured in wide dynamic range. Then, basing one of criterias, sampling frequency can be chosen. We suggest to use the A/D converter dynamic range for maximal spectra frequency calculation.