Machine Learning based Decoding using Transformer Networks

Tutor: Matthias Hummert
Type of Thesis: Project (MSc)
date of issue: 06/2023
Student: Julian S√ľndermann
Status: in progress


Recent results show that the application of Transformer Neural Networks while exploiting code knowledge can outperform State of the Art Decoder. Therefore a

deepdive into these types of decoders shall be done in this thesis to get a first understanding.


The goal of this project is to implement a Transformer neural network and perform a performance and complexity analysis for different codes.


In order to process this thesis, knowledge of the lectures Channel Coding, Advanced Channel Coding, Wireless Communications Technologies, Communication Technologies and

programming skills in Python or Matlab are essential.

Last change on 31.05.2023 by M. Hummert
AIT ieee GOC tzi ith Fachbereich 1
© Department of Communications Engineering - University of BremenImprint / Contact