Phaseless antenna characterization: efficient spectral representation and non-uniform spatial sampling

To be fruitful for wide-ranging antenna types and in different measurement conditions, phaseless antenna characterization and diagnosis algorithms must be reliable and accurate. To attain robustness against ill-conditioning in order to reach an accurate solution, the antenna characterization algorithm proposed in this paper exploits “efficient” representations of both unknowns and data. Such representations consist of a sampling series expansion of the unknown antenna plane wave spectrum and a non-uniform sampling of the squared amplitude of the near-field. In either instances the representation takes advantage of the available knowledge about the finite dimensions of the radiating system. Numerical tests against synthetic data generated by the software GRASP8-SE point out the reliability and accuracy of the proposed algorithm