ANN Performance for the Prediction of High-Speed Digital Interconnects over Multiple PCBs

Schematic of Procedure including the setup of Backplane and Daughtercard


In this paper the performance and the accuracy of artificial neural networks for the prediction of high-speed digital interconnects up to 100 GHz on printed circuit boards are analyzed and evaluated. The prediciton accuracy is evaluated both for scattering parameters in frequency domain as well as weighted power sums thereof. The interconnects considered all contain a backplane connected to a daughtercard, showing two via arrays each. Several parameter variations of the basic setup lead to a wide range of possible transmission and crosstalk parameters. Training data sets are obtained using physics-based via modeling up to 100 GHz. Approximately 7000 data sets were made available in total for this study. Neural networks are able to predict the overall link behavior.

[EPEPS 2020 - 29th Conference on Electrical Performance of Electronic Packaging and Systems]
Cheng Yang

Computational Electromagnetics, Elektromagnetische Verträglichkeit (EMV), Signal Integrity (SI) und Power Integrity (PI), Mikrowellen-Messtechnik, Maschinelles Lernen für EMV-Engineering

Christian Schuster
Professor, MLE-Gründer & -Advisor