Xueyun Long

This thesis tackles one of the hot topics for future 5G/6G networks: how to densify wireless infrastructure in cities without turning them into electromagnetic field (EMF) exposure hotspots. In dense urban environments, macro base stations (BSs), small cells, reconfigurable intelligent surfaces (RIS) and even UAV-mounted BSs are all candidates for boosting data rates for wireless communication – but they also contribute to the overall EMF exposure experienced by end-users. The goal of this thesis is to design and analyze EMF-aware deployment and configuration strategies that satisfy regulatory limits while still delivering high data rates and good coverage in a realistic urban scenario.

For this purpose, the student will work on formulating and solving optimization problems, which includes implementing and comparing different algorithms e.g metaheuristics or reinformcement learning (RL) algorithms. The work will be grounded in a realistic urban environment with building layouts, existing 4G sites, and user distributions, providing results that are directly applicable to real-world network planning.

This thesis is ideal for students interested in wireless network planning, EMF-aware 5G/6G system design, optimization algorithms, maschine learning or RL. A background in communication systems and basic signal processing is recommended.

Requirements:

• Communications engineering / signals and systems
• Interest in optimization and algorithmic problem solving
• Background of RL or ML (preferably with Pytorch)
• Knowledge of MATLAB