Company Core Technologies

Integrated Circuits, Industrial and Power Electronics

Our research for this technology field spans everything from AI-driven EDA, IC/PCB design, and manufacturing to sustainability for software-defined hardware in electrification, automation, and digitalization, integrating electronic hardware, software and mechatronic systems.

Explore publications

Imagine if AI, digital twins and software-defined hardware required less hardware and energy

What if industrial intelligence existed directly where physical work happens? As factories, manufacturing systems, and energy grids demand greater autonomy, relying on heavy server-class or cloud computing results in latency, power consumption, and hardware footprint bottlenecks. The future of automation requires intelligence to move closer to the machine edge. This approach uses a fraction of the physical hardware and offers vastly superior energy efficiency.

What matters are solutions for more efficient computing, sensing, power electronics and embedded AI capabilities for industrial systems.

A split image with a blue digital microchip visualization and three researchers looking at a laptop screen together

A robotic arm precisely assembling printed circuit boards along an automated factory conveyor belt.

Integrated Circuits, Industrial and Power Electronics encompasses advanced hardware architectures for digital, analog and mixed-signal systems, printed circuit boards and general-purpose or application-specific computing, including artificial intelligence and security workloads.

Design processes leverage machine learning and generative artificial intelligence for hardware design and verification. They cover integration, miniaturization, manufacturing processes and the full lifecycle of sustainable products. By utilizing digital thread and digital twin technologies, domain-specific systems drive hardware and software codesign across electronic platforms.

This Company Core Technology covers hardware for computing, communication, sensing, actuation and mechatronic systems to enable secure, sustainable field and edge devices for electrification, automation and digitalization. It also spans AI-enabled electronic design automation for integrated circuits and printed circuit boards, integrating digital twin and digital thread solutions across functional, electrical, thermal and mechanical domains.

Furthermore, this Company Core Technology encompasses application-specific integrated circuits for software-defined hardware, power electronics for energy conversion and motion control and hardware-aware, artificial intelligence-enabled control technologies for real-time industrial automation and robotics.

Icon for Integrated Circuits, Industrial and Power Electronics

Publications

Explore our featured papers

Spring reverb emulation with hybrid gated convolutional networks and state space models

Published in: ICASSP 2026
M. Wess, J. Janser, D. Dallinger, J. Janser, M. Wess, M. Bittner, D. Schnöll, A. Jantsch

Modeling analog spring reverbs is challenging due to nonlinear behaviors and long tails. We propose GCN-SSM, a hybrid deep learning model combining gated convolutional networks and state space models. Achieving top-tier, industry-leading perceptual quality, it requires only 125.7k parameters and allows modern CPU deployment.
View at publisher's page

Analyses of materials behavior and solder fatigue test and prediction results for long-term thermal cycling

Published in: 2025 26th International Conference on Thermal, Mechanical and Multi-Physics Simulation and Experiments in Microelectronics and Microsystems (EuroSimE)
Peter Fruehauf, Andreas Weigert, Rainer Dudek, Kerstin Kreyßig, Matthias Prinz, Susana Richter-Trummer, Sven Rzepka

This study validates FE simulations against 13-year thermal tests to predict fatigue life in FBGA277 and QFN components. Accounting for material aging and cure shrinkage in molding compounds is essential, omitting shrinkage flips calculated warpage. Aged Anand creep laws match physical failure patterns, unlike unaged data.
View at publisher's page

Reconfigurable intelligent surface for industrial automation: mmWave propagation measurement, simulation, and control algorithm requirements

Published in: 2024 IEEE 35th International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC)
L. W. Mayer, A. Hofmann, M. Schiefer, H. Radpour, M. Hofer, D. Löschenbrand, T. Zemen

Our research presents an active 127-element reconfigurable intelligent surface (RIS) operating at 23.8 GHz bypasses line-of-sight blockages in industrial mmWave environments. Controlled via FET-based amplification, it effectively focuses signals onto target points. Measurements confirm beamwidth performance, providing practical update criteria for mobile indoor tracking.
View at publisher's page

Modelling of SiC and GaN transistors based on pulsed S-parameter measurements

Published in: December 2025, Power Electronic Devices and Components
Martin Hergt, Bernhard Hammer, Martin Sack, Lukas W. Mayer, Sebastian Nielebock, Marc Hiller

In this research project, we derived a precise 12-element model for SiC MOSFET and GaN HEMT power transistors using pulsed S-parameter measurements from 2 MHz to 500 MHz. Covering pinch-off, ohmic, and active regions, the iteratively adjusted model exhibits an excellent match with original measurements for fast inverters.
View at publisher's page

AI-based motion control with system dynamics flexibility

Published in: 2024 Energy Conversion Congress & Expo Europe (ECCE Europe)
Niklas Körwer; Martin Bischoff; Michael Laumen; Rik W. De Doncker

The algorithms in today’s production machines are becoming more complex every day in order to keep up with market demands for increased productivity and flexibility. This paper proposes using reinforcement learning and neural networks to accelerate motion control development for production machines. Trained on digital twins, the network adapts motor setpoints to frequency converter dynamics, reducing following errors, smoothing control and enabling flexible reuse across various machine configurations.
View at publisher's page