A Time Skew Calibration Technique With Fast Convergence for Time-Interleaved ADCs Based on Windowed Autocorrelation Algorithm

A background calibration technique with fast convergence speed for timing skew in time-interleaved (TI) analog-to-digital converters (ADCs) is proposed in this brief. The proposed method applies a windowing technique to the results of the autocorrelation, effectively reducing the autocorrelation error caused by finite samples, and thus achieving higher accuracy in skew mismatch measurement. Compared to the conventional autocorrelation-based calibration technique, this technique significantly improves the convergence speed while only slightly increasing the hardware complexity. The simulation results show that after time skew calibration the SFDR and SNDR are improved from 55.21dB and 53.84dB to 92.67dB and 64.64dB. Finally, we implement the algorithm on a Xilinx FPGA board and utilize it as the calibration module to a 12bit 3GS/s 4-channel commercial ADC chip. The measurement results show that with the proposed technique, the SFDR is improved by 9.67dB.