Active Expansion Sampling of Magnetic Near-Fields in Unbounded Regions

Abstract

We present a fast and accurate measurement technique for magnetic near-fields by employing an adaptive sampling method with unbounded input space. The proposed machine learning algorithm is tested against uniform sampling on a printed circuit board test structure and a buck converter. We demonstrate allocation of multiple, separated regions with predefined lateral field limits at MHz frequencies. Compared to uniform sampling, active expansion sampling identifies contours of given field limits in less than 3% of the reference measurement time.