/Machine learning based beam alignment for (sub)-THz wireless communications

Machine learning based beam alignment for (sub)-THz wireless communications

Master projects/internships - Leuven | More than two weeks ago

You will be architecting the High-Speed, Ultra-Reliable and Low-Latency wireless networks of tomorrow! 

Motivated by the new emerging applications, the future (sub) THz wireless connectivity landscape of 6G and beyond will feature a wide range of applications with very-high-throughput requirements. Such high throughput can be achieved through robust signal processing to deal with the channel impairments including beam alignment and multiple access interference.  
To harvest the largest possible beamforming gain in   (sub) THz communication systems, robust collaborative transmit and receive beam alignment schemes are needed to align the beams along the dominant propagation paths between the base station and the mobile terminals.   State of the art beam alignment is amongst others heading towards hiring ideas from machine learning to replace the expensive expert-knowledge based schemes with a well-trained deep neural network, especially, for devising scalable solutions.  In this MS thesis, after a literature review, the focus will be shifted to the main competitive expert-based beam alignment techniques that will be benchmarked with machine learning solutions or hybrid schemes.  Depending on the progress, the proposed beam alignment will be evaluated subject to different beamforming architectures and their robustness will be assessed  against to channel changes and/or transceiver non-idealities.
The successful candidate must show a strong understanding of optimization, signal processing and machine learning in wireless communications.
The successful Master’s candidate will be part of a large IMEC team working on the research, implementation, and prototyping of future communications systems. The candidate should be enrolled in Electrical Engineering Master’s or a relevantly related area with a signal processing background.


Type of Project: Combination of internship and thesis 

Master's degree: Master of Engineering Technology 

Master program: Electrotechnics/Electrical Engineering 

Duration: 6 months 

Supervising scientist(s): For more information on this topic, please contact Yigit Ertugrul (yigit.ertugrul@imec.be) and Mamoun Guenach (mamoun.guenach@imec.be) 

Who we are
Accept marketing-cookies to view this content.
Cookie settings
imec's cleanroom
Accept marketing-cookies to view this content.
Cookie settings

Send this job to your email