Source current reconstruction method and field transformation with LSMR iterative solver

Abstract

The source current reconstruction method (SRM) is based on the equivalent principle in which the equivalent current of the antenna under test (AUT) is reconstructed from the known electric field and can be used for both field transformation and antenna diagnostic applications. In this method, the equivalent currents at the AUT aperture are discretized based on the known basis functions and then these currents are determined with the inverse method of moments (MoM) techniques. The large matrix equation that is created by the SRM is conventionally solved by different types of conjugate gradient (CG) solver. In this paper, a recently proposed least square iterative solver, LSMR, is used as a SRM matrix solver. Results of current reconstruction and field transformation with LSMR are reported in the paper and compared with the conjugate gradient solver. It is shown that in the presence of white Gaussian noise with different SNR, LSMR iterative solver is more stable and can reconstruct the equivalent currents sooner and better than the conventional CG solver.