Inspection tools for Gaussian Process Regression Modeling of Electromagnetic Fields of Electronic Boards and Chips

Gaussian Process Regression (GPR) has emerged as a powerful technique for building surrogate models of electromagnetic field scans in electronic systems. However, standard GPR models often struggle to capture the non-isotropic and complex spatial patterns commonly found in electronic board field distributions. In this paper, we propose some inspection tools to analyze the underlying correlation structure in the measurement data. This analysis is demonstrated by considering as input data the near field emissions from a bent miscrostrip line (full-wave simulations) and a chip (near-field measurements).