Research of a calibration method for static nonlinearity errors in TIADC

Abstract In order to solve the problem of non-ideal properties in a high-resolution Time-Interleaved Analog-to-Digital Converter (TIADC) system, various mismatch errors of the TIADC system need to be calibrated. In this paper, a single pre-Sample-Hold Amplifier (SHA) is used to avoid the timing mismatch error of this system, and a Back Propagation Neural Network (BPNN) calibration method for the static nonlinear error of this system is proposed, which fits nonlinear errors to achieve calibration by training the network parameters. The 4-channel 24bit, 10MSPS TIADC system is calibrated and simulated, this method can increase the Signal-to-Noise Ratio (SNR) of the system by more than 70 dBc. This error calibration method can improve the accuracy of the TIADC system.