INNOVATIONS UNPLUGGED – the Rohde & Schwarz technology podcast · Rohde & Schwarz

Radar reinvented? How to fast-track road safety, airport security, and AI training – all at once

·25 min·1 clip
Radar sees reflectivity and motion, not the shapes people expect from visible light.
1. Innovations Unplugged by Rohde & Schwarz centers on hyper-realistic radar simulation and how it connects radar, AI, and real-world development. 2. Markus Haller and Inga Müller-Siedentop host the episode, and guests Marcel Hoffmann and Michael Stelzig co-founded 5D, the company behind the simulation software. 3. The episode asks how radar development can move faster without losing realism, especially when engineers need labeled data and safe testing environments. 4. Michael Stelzig says radar touches automatic toilet flushing, microwave body scanners, air surveillance, aerospace defense, autonomous driving, and smart-home systems. 5. He also points to automatic cruise control as a radar application that measures distance and velocity from other cars, pedestrians, buildings, and guardrails. 6. Marcel Hoffmann says 5D does not build radar hardware, but its simulation solution helps test radar systems, software, and sensor concepts before physical prototyping. 7. He traces the company’s origin to a university pain point, where a PhD project could not realistically move a radar concept from development to street testing inside a thesis timeline. 8. The guests say the same timing problem affects large automotive companies, because prototyping, data gathering, and integration into vehicles can take years. 9. They use drone detection at Munich airport as an example of a situation where real-world testing is constrained by operations, flight limits, weather, and sensor types such as cameras. 10. They describe a second gap in simulation: low-level physics models are too computationally expensive, while high-level path-planning simulations are not realistic enough for radar development. 11. Michael Hoffmann says radar should not be thought of like visible light, because radar relies more on reflectivity and motion than on shapes and colors. 12. He compares radar interpretation to ultrasound imaging, where the raw picture is less intuitive than the clinician’s explanation of what the sensor sees. 13. Markus Haller says a pedestrian returns not a single dot but a set of motion traces, including torso movement and swinging arms, which surprised him as a physicist. 14. Marcel Hoffmann names that motion pattern a micro Doppler signature and says it can act like a fingerprint for classifying moving objects. 15. The guests say AI needs large, labeled datasets for radar, and simulation can generate those datasets by varying pedestrians, speeds, and scene contents with a click. 16. Marcel Hoffmann says 5D’s USP is a processing chain that builds digital radar twins with hardware effects, non-linearities, and wave propagation realism to reduce the domain gap. 17. Christina Gessner says Rohde & Schwarz values collaboration with universities and research because it turns ideas from customer needs and technology potential into real products. 18. The episode’s style is technical but conversational, with the hosts asking for plain-language explanations and the guests using concrete examples from airports, cars, and research. 19. Listeners interested in radar, automotive sensing, airport security, or AI training data will get the most from it. 20. Listeners looking for a light general-audience chat or non-technical banter will probably skip it.

As heard by us

A clear look at radar simulation, motion signatures, and the AI training that helps make them usable.

Radar is framed as something that reaches into ordinary life more often than many people notice, from automatic toilet flushing to more advanced sensing systems.

Read the full review in PlayNext →

Why you'd press play

Want radar explained through cars, pedestrians, and AI training data?

Read the full recommendation in PlayNext →
Listen to the show on