Master Thesis on Federated Learning for Internet of Things (IoT) Systems

German Aerospace Center (DLR); Oberpfaffenhofen; Vor Ort; EUR 40.000 - 60.000; Vollzeit

Zusammenfassung

The German Aerospace Center (DLR) is offering a thesis opportunity in Oberpfaffenhofen focusing on federated learning solutions for Internet of Things applications. The role involves studying various parameters of NB-IoT pertaining to MAC protocols using ns-3 simulations. Ideal candidates should have programming skills in C/C++, knowledge of communication systems, and an interest in machine learning. The position entails independent work and exploration of interdisciplinary topics, contributing to state-of-the-art research.

Qualifikationen

• Good knowledge of communication systems and signal processing, with an emphasis on MAC level protocols.
• Interest in machine learning and federated learning.
• Programming skills in C/C++ and willingness to learn new tools.
• Previous experience with event‑driven network simulations is a plus.
• Ability to work independently and interest in interdisciplinary topics.
• Good knowledge of the English language.

Aufgaben

• Familiarize with 3GPP NB-IoT aspects at the MAC layer.
• Familiarize with the ns-3 network simulator.
• Study FL algorithms performance over NB-IoT using ns-3 simulations.
• Optimize NB-IoT protocol parameters for practical use-cases.

Kenntnisse

Knowledge of communication systems

Interest in machine learning

Programming skills (C/C++)

Experience with event-driven network simulations

Good knowledge of English

Tools

ns-3 network simulator

Jobbeschreibung

The DLR Institute of Communications and Navigation is dedicated to mission‑oriented research in selected areas of communications and navigation. Its work ranges from the theoretical foundations to the demonstration of new procedures and systems in a real environment and is embedded in DLR's Space, Aeronautics, Transport, Security and Digitalization programmes.

What To Expect

The Advanced Information Processing Group aims at applying state‑of‑the‑art theoretical results into real‑world applications within information processing systems. The expertise of the group ranges from quantum error correction to Smart Data Management, exploring cutting‑edge communication theories such as semantic communication and Age of Information, pushing the boundaries of data utilization and dissemination.

The thesis will focus on the study and optimization, by means of network simulations, of federated learning (FL) solutions for Internet of Things (IoT) applications. Specifically, the possibility to efficiently support FL over 3GPP narrowband‑IoT (NB‑IoT), a core element of IoT connectivity in 4G, 5G and 6G systems, will be explored, with attention to MAC protocol aspects.

Your tasks

• Familiarize with the main aspects of 3GPP NB‑IoT at the MAC layer
• Familiarize with the ns-3 network simulator
• Study, by means of ns-3 simulations how FL algorithms perform over NB‑IoT
• Optimize the NB‑IoT protocol parameters to efficiently support FL in use‑cases of practical relevance

Your profile

• Good knowledge of communication systems and signal processing, with particular emphasis on MAC level protocols
• Interest in machine learning and federated learning
• Programming skills (C/C++) and willingness to learn new tools
• Previous experience with event‑driven network simulations is a plus
• Ability to work independently and interest in interdisciplinary topics
• Good knowledge of English language