Blind calibration method for two-channel time-interleaved ADCs based on cross correlation and polynomial regression

Time-interleaved analog-to-digital convert (TIADC) is an effective method to improve the sampling rate of ADC system. However, timing mismatch in the TIADC system degrades the spurious-free dynamic range (SFDR) of the TIADC system greatly. In this paper, a blind calibration algorithm based on cross correlation and polynomial regression is proposed, which can estimate and calibrate the timing mismatch blindly. The proposed algorithm needs no extra circuits and no training signals, it can dynamically track the changes of timing mismatch. Simulation and a 130MS/s two-channel TIADC prototype are used to verify the proposed algorithm. The improvement of SFDR is 15dB in average in the frequency band of [0.1 π, 0.8 π]. The simulation and experiment results show that the proposed algorithm has improved the performance of TIADC system effectively.