LEARNING OUTCOMES
Communication system design has traditionally relied on developing a mathematical model and producing optimized algorithms for that model. However, with the increasing access to data and computing resources, a data-driven approach based on machine learning has gained interest in recent years. This course provides a brief introduction to machine learning that is tailored for communication and information theory researchers. Key concepts of machine learning will be introduced and exemplified with applications in communications.
Credits: 5
Schedule: 10.01.2025 - 10.04.2025
Teacher in charge (valid for whole curriculum period):
Teacher in charge (applies in this implementation): Hanan Al-Tous
Contact information for the course (applies in this implementation):
CEFR level (valid for whole curriculum period):
Language of instruction and studies (applies in this implementation):
Teaching language: English. Languages of study attainment: English
CONTENT, ASSESSMENT AND WORKLOAD
Content
valid for whole curriculum period:
This course consists of the following modules:
Module I: Channel Models:
Machine learning methods are data hungry and to obtain reliable results huge amount of real world measurements are needed. This problem can be circumvented by synthetic data which mimics the behavior of realistic channels. Channel models appropriate for Fifth Generation (5G) and Beyond Fifth Generation (B5G) systems will be studied. Channel model characteristics including path loss, large-scale fading, small-scale fading models and spatial consistency will be discussed. The differences of Milimeter-Wave and microwave models will be explained. Several radio channel simulators will be introduced.
Module II: Massive MIMO and Beam Management:
Massive multiple-input multiple-output (MIMO) is an important technology in 5G and B5G systems. Beam management procedures are used to acquire and maintain a set of beam pair links (a beam used at gNB paired with a beam used at user equipment). These procedures aim to maintain high-quality communication links despite challenges like path loss, blockages, and rapid changes in user equipment position and orientation. We will discuss some use cases of applying machine learning for beam management in future communication systems.
Module III: Radio-frequency Positioning:
Location awareness is essential for enabling location based services and for improving network management in future communication systems. We will discuss the use of radio frequency based approaches to localization. We review the radio frequency features and techniques that can be utilized for positioning. We will discuss the challenges for indoor positioning and utilize machine learning techniques to address these challenges.
Module IV: Channel Charting and Applications:
Channel charting is a self-supervised machine learning framework, that is applied to the channel state information of the users in wireless systems to create a logical radio map of radio environment. The channel chart can be then used for several radio resource management applications. We will study dimensionality reduction techniques that can be used for channel charting and discuss several radio resource management applications based on channel charting.
Module V: Neural Network Structures and Applications
We will study neural network structures and concepts to jointly optimize the transmitter and receiver in communication system by a single process. We will utilize the autoencoder structure to represent the end-to-end communication system.
Assessment Methods and Criteria
valid for whole curriculum period:
Exercises and Examination
Workload
valid for whole curriculum period:
Contact teaching and indvidual studies. Teams work etc.
DETAILS
Substitutes for Courses
valid for whole curriculum period:
Prerequisites
valid for whole curriculum period:
FURTHER INFORMATION
Further Information
valid for whole curriculum period:
Teaching Language: English
Teaching Period: 2024-2025 Spring III - IV
2025-2026 Spring III - IV