This article deals with the relatively late introduction of entertainment radio in Germany compared to other technically advanced countries. Following current broadcasting history, the German Post Office was initially interested in the commercial use of wireless technology after the end of the First World War. Due to the success of entertainment radio in the USA and UK, there were demands from industry and radio amateurs for its opening in Germany as well, which finally happened with a delay.

This paper aims to classify electromagnetic compatibility (EMC) issues using autoencoders, a dimensionality reduction technique, and machine learning models. The process begins by generating EMC measurement data that closely reflects real-world measurements. The samples are then reduced using autoencoders and used as input for the machine-learning models. The results demonstrate that the machine learning techniques were able to accurately classify between the different EMC classes. 

In electronic design development, schematics are used to depict electrical circuits and validate them before manufacturing. As such designs are often handed in as images (bitmaps) only, there is no way to easily process and check them in an automated way.
With the help of deep-learning methods, components, lines and text can be extracted from these bitmaps.
Fusing these detections, the electrical circuit can be reconstructed and serve as an input for further validation routines.