Automatic sparse ESM scan using Gaussian process regression

Abstract

Emission source microscopy (ESM) technique can be utilized for localization of electromagnetic interference sources in complex and large systems. In this work a Gaussian process regression (GPR) method is applied in real-time to select sampling points for the sparse ESM imaging using a motorized scanner. The Gaussian process regression is used to estimate the complex amplitude of the scanned field and its uncertainty allowing to select the most relevant areas for scanning. Compared with the randomly selected samples the proposed method allows to reduce the number of samples needed to achieve a certain dynamic range of the image, reducing the overall scanning time. Results for simulated and measured 1D scans are presented.