Fast Nonlinear Reconstruction for Radiation Patterns With Phaseless Measurement in Reverberation Chamber

This paper explores the potential of a nonlinear optimization algorithm, known as the Levenberg-Marquardt (LM) algorithm, to expedite the calculation of spherical wave coefficients. Reconstruction of antenna radiation patterns (AUT) within a reverberation chamber (RC) could be accomplished through the utilization of spherical wave expansion (SWE) with the measurement of power correlation coefficients. However, solving the SWE of electrically large AUT is complex and time-consuming. The matrix decomposition approach is used to convert high-dimensional matrices into multiple low-dimensional ones. Additionally, it analytically solves the Jacobian matrix within the LM, achieving a time savings of more than fivefold. The efficacy and precision of this method are substantiated through numerical calculations, simulations, and measurements.