Semantic Communication System Design via Information Bottleneck Method

Tutor: Edgar Beck
Type of Thesis: Project (MSc), Master's thesis (MSc)
date of issue: -
Student: -
Status: available
Abstract:

Motivation:

Motivated by recent success of machine learning tools at the PHY layer and driven by high bandwidth demands of the next communications standard 6G, the old idea of semantic communication by Weaver from 1949 has received considerable attention. It breaks with the classic design paradigm according to Shannon by aiming to transmit the meaning of a message rather than its exact copy and thus potentially allows for savings in bandwidth. In our first work, inspired by Weaver, we proposed an information-theoretic framework where the semantic context is explicitly introduced into probabilistic models. In particular, for bandwidth efficient transmission, we define semantic communication system design as an Information Bottleneck optimization problem and considered important implementation aspects. Further, we uncovered the restrictions of the classic 5G communication system design w.r.t. semantic context. Notably, based on the example of distributed image classification, we revealed the huge potential of a semantic communication system design. Numerical results show a tremendous saving in bandwidth of 20 dB with our proposed approach ISCNet compared to a classic PHY layer design.

Goal:

The aim of this thesis is first to become familiar with the research domain of semantic communication and then to understand our new proposed approach ISCNet. Then, you implement ISCNet - but this time with an explicit instead of an implicit Information Bottleneck.

Requirements:

In order to work on this thesis, a successful participation in the lectures Wireless Communications and Communication Technologies is required. Also, a solid understanding in linear algebra, optimization theory and stochastics is needed. Programming skills in Python and participation in the lecture Advanced Topics in Digital Communications are advantageous but not necessary. In order to work on this thesis, a successful participation in the lectures Wireless Communications and Channel Coding is required. Also, a solid understanding in linear algebra, optimization theory and stochastics is needed. Programming skills in Python/Tensorflow and participation in the lecture Advanced Topics in Digital Communications and/or Information Theory are very advantageous but not necessary.

Last change on 26.04.2022 by E. Beck
AIT ieee GOC tzi ith Fachbereich 1
© Department of Communications Engineering - University of BremenImprint / Contact