| 摘要 |
A system and method for reducing linearity errors in an A/D converter, such as a delta-sigma converter. The linearity errors in the delta-sigma converter are modeled by generating a set of digital signals representative of an inputted sine wave. The set of digital signals are low-pass filtered and subjected to a fast Fourier transform algorithm to generate a frequency domain representation of the sine wave. Thereafter, a net linearity error spectrum is removed from the frequency domain representation and inverse Fourier transform back into the time domain. The filtered set of digital signals are also sorted into subsets of digital signals where each signal in a subset corresponds to a particular output of a delta-sigma modulator contained within the delta-sigma converter. A fast Fourier transform algorithm is applied to each of the filtered subsets of digital signals to generate a frequency domain representation thereof. Specific linearity errors are generated by applying an inverse Fourier transform algorithm to each of the specific linearity error spectrums in the frequency domain representations of the filtered subsets of digital signals. Thereafter, linearity error correction coefficients are generated as a function of the net linearity error and the specific linearity errors. The linearity error correction coefficients are used to generate entries in a look-up table where the entries are adjustable by digital outputs of the delta-sigma modulator. The look-up table is used to correct digital signals outputted by the delta-sigma modulator prior to decimation and digital filter.
|
