Evaluation of Simulated and Real Measurement Data for AI-based Interference Classification in EMC Applications

This paper aims to compare real-world and simulated measurement data of printed circuit boards (PCBs) and identify relationships between input and output classes. The data is evaluated using dimensionality reduction techniques and feature-to-output correlation metrics. This analysis aims to provide insights into the accuracy and reliability of simulated data in representing real-world PCB measurements. The goal is to enhance simulation accuracy and reliability in PCB design and manufacturing processes.