Efficient Near-Field Recovery from Non-Redundant Cylindrical Near-Field Measurements

Abstract

This work presents an efficient method for reconstructing the field components on an arbitrary near-field (NF) cylindrical surface external to the antenna under test (AUT), starting from non-redundant (NR) voltage cylindrical samples. Unlike conventional approaches, which, for each cylindrical NF component and output point, require the evaluation of as many integrals as the number of azimuthal harmonics, the proposed method directly formulates a NR representation on the reconstruction surface, thus minimizing the number of integrals to be evaluated. At the same time, by exploiting the NR sampling theory, it also minimizes the amount of measurements needed on the scanning cylinder. A two-dimensional optimal sampling interpolation is then used to get the NF components at any point on the target surface, thus drastically reducing the computational cost. Numerical results confirm the accuracy of the method.