Influence of algorithm parameters on static nonlinearity in switching-based calibrations for DACs

Circuits and Systems for Receiving, Transmitting and Signal Processing

This paper presents a comparative study of nonlinearity reduction by different algorithms of switching-based calibration method for DACs. Based on the known algorithms, one general parametric algorithm is proposed. The introduced parameters are a resolution of primary array and numbers of folding and decoupling operations. The intermediate options of such a general algorithm, defined by the parameters combination, are called “algorithm cases”. The algorithm cases are compared by the efficiency of static nonlinearity reduction in presence of random and systematic errors. For both errors, the folding operation improves DNL in contrast with the decoupling operation, which deteriorates it. The increment of folding operations number lowers the presence of systematic drift in DNL, which completely disappears at the maximum number. In the case of INL, the impact of both operations is determined by the elements order. If elements values depend on their indices, the folding operation deteriorates INL in contrast with the decoupling operation, which improves it. All fully unarized arrays provide the INL reduction, but do not decrease DNL.