Precise wideband millimeter signal generation with machine learning techniques

07 May 2020

Get in touch!


For more information please contact Prof. dr. ir. Dominique Schreurs, mail:

You can apply for this job no later than May 21, 2020 via the online application tool

KU Leuven seeks to foster an environment where all talents can flourish, regardless of gender, age, cultural background, nationality or impairments. If you have any questions relating to accessibility or support, please contact us at

Ref. BAP-2020-302

Apply before 21 May 2020

The Division ESAT-TELEMIC is a multidisciplinary research group of the Department of Electrical Engineering (ESAT) focusing on a broad range of topic areas ranging from electromagnetic theory, microwave/millimetre wave devices and circuits, to telecommunication systems. This PhD vacancy is in the team of Prof. D. Schreurs, which long-standing experience in microwave/mmwave experimental techniques for applications ranging from wireless communications to biomedical diagnostics.


With demands for ever higher data rate wireless communications and very wideband sensing, wideband signals at carrier frequencies in the millimeter wave frequency range are being targeted. An emerging evolution is to adopt frequency multipliers (freqX) in the transmitter chain, as to boost the carrier frequency strongly with minimal radiofrequency electronics as compared to the traditional mixer-oscillator stage. Two goals are targeted in this project: (objective A:) the characteristics of the input signal are to be preserved at the freqX output (application: wireless communications), and (objective B:) a predefined wideband excitation is to be generated precisely at the freqX output (applications: radar and metrology).Regarding objective A, the main challenge is to conceive a proper signal predistortion methodology, considering the strong intrinsic nonlinearity order of freqX. By in-depth forward and inverse modelling, the project targets to achieve results for signals with bandwidths an order of magnitude larger than the state-of-art.Regarding objective B, the idea is to exploit intentionally the bandwidth expansion property of freqX's nonlinear operation as to achieve precisely predefined wideband signals. In addition, preserving signal quality when loading the freqX by the next block in the transmitter, will be investigated for the first time. Throughout the project, there is strong emphasis on real-life proofs-of-concept with extensive measurement validation.


MSc in electrical engineering, with background in microwave engineering and strong interest in machine learning, or vice versa.


PhD scholarship for 4 years, including pre-doc phase if required.