forum | IKS - Details
forum | IKS: Vortrag von Prof. Dr.-Ing. Laurent Schmalen
Karlsruhe Institute of Technology
Improving Communication Systems Using Machine Learning
Freitag, 05. März 2021
virtueller Konferenzraum - Einwahldaten auf Anfrage
Im Rahmen der IKS‐Veranstaltungsreihe referierte Herr Prof. Dr.-Ing. Laurent Schmalen, Karlsruhe Institute of Technology (KIT) über das Thema:
"Improving Communication Systems Using Machine Learning"
Today, communication engineering still follows a model-based design methodology influenced by the seminal design guidelines that were formulated by Claude Shannon in the 1940s. Such a model-based approach may however not be suitable for many modern communication scenarios. In this talk, we show how we can augment communication systems using machine learning and in particular deep learning.
In the first part of the talk, we show how machine learning can be used to optimize channel-agnostic waveforms for an optical communication system. In the second part of the talk, we show how machine learning can be used to augment existing receiver algorithms and in particular channel decoding. We illustrate that short channel codes can be decoded with a performance close to the theoretical performance limits with significantly lower complexity than other state-of-the-art methods.