The expansion of information networks such as the Industrial Internet of Things (IoT) is creating an ever-increasing demand for data. As a logical consequence, this creates a demand for data generation, processing, storage and data transport between these steps. This project is investigating ways of ensuring fast data transport. The aim is to use machine learning (ML) methods to accelerate the design process for high-speed connections. The main focus is on the structure of the PCB, but the effects of line coding and equalizers are also considered. The overall aim is to create a large SI/PI database on the basis of which a hybrid design flow can be created that combines data-driven and physics-based methods. It will be investigated for which sub-areas and in which way the respective ML methods can be used optimally.

Funding: Freie und Hansestadt Hamburg
Contact: Til Hillebrecht, M.Sc.
Start date: 17.10.2022

In the current market, there is a growing demand for enhanced power integrity design in modern integrated circuits, driven by increasing data rates and lower supply voltages. The design cycle involves numerous electromagnetic simulations, which require substantial time and computational resources. The application of machine learning offers a potential speed enhancement for accurate simulation results. Additionally, the combination of physical knowledge and data-driven approaches can provide novel insights into the behavior of printed circuit boards. The rise of transformer architectures and the resulting success of large language models is investigated regarding their applicability in electronic designs. The goal of the project is to establish a machine learning assisted workflow for the design of printed circuit boards regarding the compliance to power integrity requirements.

Funding: Freie und Hansestadt Hamburg
Contact: Jan Heßling, M.Sc.
Start: 03.03.2025

High-speed interconnect design for wired links in digital systems has been a well-established field of research and innovation for several decades now. In contrast, automotive high-speed interconnects have only recently advanced to higher and higher data rates. Correspondingly, models for components specific to automotive interconnects have not been created or validated to a large extent nor has a sufficient model-to-hardware correlation been established. In addition, automotive high-speed interconnects must fulfill strict requirements regarding electromagnetic compatibility (EMC) regulations. This project aims to develop high-speed interconnect models used in automotive applications on a system level. These models will be adapted to analyze printed circuit board (PCBs) stack-ups used in automotive applications in combination with assembly and interconnection technologies (AITs) and provide insight of the Signal Integrity performance of high-speed links in harsh automotive environmental standards to have a functional system which fulfill the electrical performance as well as reliability requirements.

Funding: Robert Bosch GmbH
Contact: Jose Enrique Hernandez Bonilla, M.Sc
Start date: 01.09.2022

Physics-based simulation tools offer an efficient alternative to full-wave solvers for predicting the signal integrity behavior of PCB interconnects. By leveraging simplified analytical models of the underlying electromagnetic phenomena, they achieve significantly lower computation times while maintaining accurate results — currently for frequencies up to 40 GHz. However, as modern high-speed applications push data rates ever higher, these simplified models reach their limits: the electromagnetic behavior becomes increasingly complex, and effects that were previously negligible can no longer be ignored. To address this, my research investigates the extension of physics-based tools to frequencies up to 100 GHz by developing refined models that capture this additional complexity — such as more accurate descriptions of the electromagnetic field behavior around via structures and their mutual interactions. Where purely analytical descriptions become impractical, data-driven or numerical methods are explored to complement the modeling approach.

Contact: Tommy Weber, M.Sc
Start: 01.04.2026