Recursive least squares adaptive digital background calibration of analog-to-digital converters

This paper proposes an adaptive digital background error-correction technique to calibrate analog-to-digital converters (ADCs). The new approach combines an adaptive recursive-least-squares (RLS) algorithm-based FIR filter and an accurate reference ADC. Matlab simulation indicates that the proposed RLS filter is sufficient to remove the effect of large differential and integral nonlinearities resulting from component errors including capacitor mismatch, finite op-amp gain, op-amp offset and sampling-switch-induced offset. With a 100MHz sinusoidal input, the ENOB can be increased from 8.56bit to 15.55bit, the peak SNR can be increased from 53.29dB to 59.35dB and the SFDR can be increased from 56.52dB to 119.26dB.