Postdoc in machine learning optimization and FPGA acceleration

at  Springer Nature

POSTDOC IN MACHINE LEARNING OPTIMIZATION AND FPGA ACCELERATION

Employer
Interdisciplinary Centre for Security, Reliability and Trust (SnT), University of Luxembourg
Location
Luxembourg
Salary
Negotiable
Closing date
4 Jun 2024
Discipline
Computing, Engineering
Job Type
Postdoctoral
Employment - Hours
Full time
Duration
Fixed term
Qualification
PhD
Sector
Academia

You need to sign in or create an account to a job.

• Job Details
• Company

ABOUT THE SNT…

The University of Luxembourg is seeking to hire a highly motivated and outstanding researcher in the area of machine learning optimization for physical layer applied to satellite communications in its Interdisciplinary Centre of Security and Trust (SnT), within the Signal Processing and Communications (SigCom) research group, led by Prof. Symeon Chatzinotas.
SnT carries out interdisciplinary research in secure, reliable and trustworthy ICT (Information and Communication Technologies) systems and services, often in collaboration with governmental, industrial and international partners. SnT is active in several national projects funded by National Research Fund (FNR) and local industries, and international research projects funded by the EU FP7 programme, H2020 programme, and the European Space Agency (ESA). For further information, you may check: www.securityandtrust.lu
The SigCom research group carries out research activities in the areas of signal processing for wireless communication systems, including satellite and 5G/6G communications. It is expanding its research activities to explore several emerging use cases of next-generation wireless communications systems. For details, you may refer to the following: https://wwwen.uni.lu/snt/research/sigcom
We’re looking for people driven by excellence, excited about innovation, and looking to make a difference. If this sounds like you, you’ve come to the right place!

This is a fully funded postdoc position for 2 years within an ESA project (TANNDEM-Demodulator based on Artificial Neural Network). Therefore, the objectives of the postdoc are defined in accordance with the TANNDEM directions toward reducing complexity and resources in the receiver, as well as increasing the performance for integrating satellites on 5G networks. More precisely, the work aims at designing artificial neural networks for different functionalities in the physical layer.
The successful candidate will work under the academic supervision of Dr. Jorge Querol (University of Luxembourg) and join a strong and motivated research team led by Prof. Symeon Chatzinotas.

The successful candidates are expected to perform the following tasks:

• Optimize 5G New Radio (NR) with waveform based on orthogonal frequency-division multiplexing (OFDM) and its evolution for satellite communication (Satcom) applications
• Design machine learning models to reduce the computational complexity and power consumption of several physical layer processing blocks of the receiver, such as decoding, channel estimation, and demodulation
• Ensure the ML models reduce the complexity while maintaining satisfactory performance in terms of throughput and bit-error-rate
• Identify the most suitable metrics to evaluate the machine learning models considering the system performance
• Implementing and developing a software and/or hardware testing environment to compare and evaluate the proposed techniques and architectures
• Disseminating results through scientific publications and conferences
• Contribute to the different research projects, either coordinating technical contributions and/or preparing project deliverables
• Preparing new research proposals to attract industrial, national, and European projects
• Eventually providing assistance in the supervision of PhD students