A calibration algorithm of millimeter-wave sparse arrays based on simulated annealing

A new calibration method developed for millimeter wave sparse arrays is presented. The method estimates a calibration matrix to compensate for unequal gain and phase responses, and the small errors in element positions simultaneously. Firstly, the data collections are done by placing two pilot sources in different near-field positions. Then taking diagonal sensor gain/phase perturbation matrix and element position errors as optimization parameters, the objective function is presented according the principle of signal subspace. Using simulated annealing algorithm, the diagonal sensor gain/phase perturbation matrix and element position errors are estimated accurately finally. The usefulness and behavior of the proposed calibration method are illustrated though simulated experiments.