An efficient method for modeling and evaluating quantization effects on Gaussian signals

The quantization of sampled Gaussian signals arises naturally in a wide range of digital signal processing scenarios. Since the quantization characteristic has many discontinuities, it is difficult to make accurate polynomial of it. Therefore, the analytical evaluation of quantization effects is in general difficult. In this paper we present a simple, yet accurate, method for the analytical study and efficient simulation of quantization effects on sampled Gaussian signals. Our approach is to define an equivalent polynomial characteristic with a relatively low number of terms that produces the same spectral effects on Gaussian signals as the quantizing characteristic. As an example, we employ our method for the performance evaluation of quantization effects on multicarrier signals. Our results show that our approach is very efficient, even for oversampled signals.