An efficient blind timing skews estimation for time-interleaved analog-to-digital converters

In this paper, we propose an efficient blind timing skews estimation (TSE) method for TIADC systems by evaluating the autocorrelation of the analog input signal and the mean squared difference between the output samples of the two adjacent sub-ADCs. By approximating the input signal autocorrelation function using the second order Maclaurin series, an analytical solution of the blind TSE using noise free sub-ADC output model and additive white Gaussian noise model is derived. Experimental results show that the proposed blind TSE method has low computational complexity, good estimation accuracy and robustness to additive Gaussian noise for the analog signals with input frequency lower than half of the Nyquist frequency. The experiment results with the real TIADC output data further validate its good performance.