AI-Driven EMC Diagnostics and Simulations for PCBs
EMC Professional Talk
AI-Driven EMC Diagnostics and Simulations for PCBs
PCBs play a critical role in shaping the electromagnetic behavior of modern electronic systems. As integration densities and operating frequencies continue to increase, PCBs are often the major sources of unwanted emissions. This makes them a central focus in EMI troubleshooting and regulatory compliance. This talk presents an overview of recent work that leverages ML to address key challenges in PCBs with respect to EMC. Data-driven approaches are utilized to detect PCB anomalies and emission sources using an automated scenario. Furthermore, Physics-informed neural networks (PINN) are employed as an efficient meshless simulation framework that is capable of simulating field distributions within multilayer PCB structures. These simulations offer significant advantages in speed and flexibility, as they avoid meshing and solve directly on scattered spatial points opening new possibilities for fast and computationally efficient electromagnetic simulation frameworks.
Date and Time
Location
Hosts
Registration
- Date: 24 Jun 2025
- Time: 12:00 PM UTC to 01:00 PM UTC
-
Add Event to Calendar
- Starts 06 June 2025 10:00 PM UTC
- Ends 23 June 2025 10:00 PM UTC
- No Admission Charge
Speakers
Mohamed Kheir of University of Southern Denmark
AI-Driven EMC Diagnostics and Simulations for PCBs
PCBs play a critical role in shaping the electromagnetic behavior of modern electronic systems. As integration densities and operating frequencies continue to increase, PCBs are often the major sources of unwanted emissions. This makes them a central focus in EMI troubleshooting and regulatory compliance. This talk presents an overview of recent work that leverages ML to address key challenges in PCBs with respect to EMC. Data-driven approaches are utilized to detect PCB anomalies and emission sources using an automated scenario. Furthermore, Physics-informed neural networks (PINN) are employed as an efficient meshless simulation framework that is capable of simulating field distributions within multilayer PCB structures. These simulations offer significant advantages in speed and flexibility, as they avoid meshing and solve directly on scattered spatial points opening new possibilities for fast and computationally efficient electromagnetic simulation frameworks.
Biography:
About the speaker:
Mohamed Kheir received his MSc degree in Communications Technology from Ulm University in 2005, and the PhD degree (with honors) in Information Engineering and Technology from the German University in Cairo, jointly with Magdeburg University, in 2011.
From 2012 till 2015, he was a Lecturer at Kiel University where he participated in teaching and research. From 2015 to 2020, he was a Senior RF/EMC Expert with IMS Connector Systems Group and Keysight Technologies. Since 2020 till 2023, he has been a Technical Leader for EMC Simulations at Nokia Networks, Ulm, Germany. Since 2023 he has been an Associate Professor and EMC Group Leader at University of Southern Denmark (SDU), Sønderborg, Denmark. Dr. Kheir serves as an Editor at IEEE Access and IEEE Internet of Things Journal, a reviewer for multiple IEEE Journals. He is a recipient of several awards and recognitions from different organizations worldwide. He also serves as an Officer for Standards Development at the IEEE EMC Society. His research interests focus on ML-Driven Techniques for EMC Simulations and Novel EMI Shielding. He is also an active member and professional at the IEEE EMC German Chapter.
Address:Denmark