Calibration of Analog to Information Converter based Sampling System

Analog to information converter (AIC) technique is an efficient approach to achieve sub-Nyquist sampling, and is also the most popular application of the compressed sensing theory. However, the parallel architecture AIC acquisition system suffers from the mismatch of channels, especially the phase mismatch. Due to the asynchronism of pseudo random sequence, it will introduce the mismatch between practical system and the analysis model, and degrades the performance of reconstruction. In this paper, we investigate the impact of gain, offset, and phase mismatch on the reconstruction, and propose a calibration method. A calibration matrix is proposed for the phase mismatch, which is constructed based on fractional delay filter. Experiment is reported to evaluate the proposed calibration algorithm, and the results indicate that the proposed approach is feasible and efficient.