AP5 Signal Processing & Algorithms

Lead: Ruhr-Universit├Ąt Bochum, Prof. Sezgin

In AP5, topics related to signal processing and algorithms are further developed. This work package covers a wide spectrum, ranging from imaging spectroscopic techniques for material characterization to real-time inline measurement methods for multi-channel systems, as well as methods for efficient resource management. Combining THz systems with, for example, imaging 3D system concepts requires new approaches to handle the resulting data volumes. To address this, sensor-based AI methods, highly parallelized decentralized signal processing methods, and neural networks need to be developed, for which there is practically no existing data foundation in the THz domain. Therefore, within the framework of this work package, various algorithms and signal processing methods are addressed, closely related to the demonstrator systems. The focus of this work package is particularly on imaging algorithms as well as resource-efficient and efficient methods of data acquisition and processing. As far as we know, none of these topics have been comprehensively addressed in the research literature for the THz frequency range.