Abstract—Modulated Wideband Converter (MWC) is a type of practical Compressed Sensing (CS) framework that bridges CS theory to real-time sub-Nyquist sampling and recovery of sparse wideband analog signal with low computational overhead and efficient hardware implementation. However, the application and implementation of MWC are seriously disturbed by the measurement matrix mismatch problem caused by limited system bandwidth and inevitable nonlinearity of practical devices. To remit this awkwardness, a Full-column-rank Well-Posed Calibration (FWC) method aiming to calibrate measurement matrix of MWC is proposed in this paper. By controlling the number and carrier frequency of input analog signal, the underdetermined reconstruction problem is turned to be well-posed equation set which is solved iteratively and then the calibration of the measurement matrix is completed. Theoretical analysis and numerical simulation results indicate that the measurement matrix obtained by proposed method keeps a high accordance with the theoretical value and adapts to the practical MWC system, thus ensuring the recovery accuracy of the original signal.

Index Terms—Compressed sensing, modulated wideband converter, measurement matrix, calibration

Cite: Tianyi Xiong, Zan Li, and Peihan Qi, “Measurement Matrix Calibration Method for Modulated Wideband Converter under Imperfect Periodic Pseudorandom Waveform," Journal of Communications, vol. 11, no. 5, pp. 498-506, 2016. Doi: 10.12720/jcm.11.5.498-506