An Antenna Array Diagnosis Approach Based on CNN Inversion and CFAR Detection

Abstract

In this article, a novel approach for the diagnosis of planar antenna arrays is presented. The developed method is based on the use of a U-Net convolutional neural network (CNN) for reconstructing the surface electric-field distribution over the array aperture starting from measurements of the radiated field collected in the far-field region. The obtained distributions are subsequently postprocessed through a constant false alarm rate (CFAR) approach to identify the possibly faulty elements. The proposed technique has been validated using numerically simulated data concerning realistic patch arrays, showing good detection capabilities.