The Influence of an Analog-to-Digital Converter on the Probability Characteristics of Gaussian Noise

Abstract

We analyze the influence of an analog-to-digital converter (ADC) on the statistical characteristics of stationary Gaussian noise with allowance for two main effects. The first effect, caused by the limited magnitude of the signal at the ADC output, is the truncation of “tails” of the probability distribution function. The second effect is the noise quantization, i.e., transformation of the continuous probability distribution function of input noise to a set of readout probabilities at the ADC output. It seems that these effects were not analyzed in the literature earlier. The efficiency of using ADC, which is the noise standard normalized to the maximum permissible amplitude of the ADC input voltage, is chosen as the parameter characterizing the effect of truncation of the tails of the probability distribution function. In addition, we vary the ADC digit capacity, which determines the noise quantization degree. We propose recommendations on the choice of the digit capacity and the efficiency of using ADC.