A statistical methodology for the design of high-performance CMOS current-steering digital-to-analog converters

With the shrinking of device sizes, random device variations become a key factor limiting the performances of high-resolution complementary metal-oxide-semiconductor (CMOS) current-steering digital-to-analog converters (DACs). In this paper, we present a novel design methodology based on statistical modeling of MOS transistor drain current that allows designers to explore different DAC architectures and to study the effects of technological variations on system performance without using time-consuming Monte Carlo simulations. This technique requires as a first step the estimation of the mean value and the autocorrelation function of a single stochastic process. This stochastic process models the device drain current and summarizes all the random sources associated with the process/device variations since the current represents the effect of all of them. Subsequently, on the basis of such an approach, a behavioral model of current-steering DACs has been developed. Finally, the statistical simulation of static performances such as differential nonlinearity and integral nonlinearity has been carried out for different DAC architectures based on the behavioral model previously derived