Future 6G wireless networks are expected to operate in highly dynamic 3D environments involving terrestrial base stations, unmanned aerial vehicles, satellites, and mobile users. In such networks, interference becomes strongly time-varying and depends on multiple interacting factors such as user mobility, altitude, beam direction, traffic load, and network geometry. This thesis investigates how causal inference can be used to improve interference prediction in 3D wireless networks. Instead of relying only on correlation-based machine learning models, the goal is to identify causal relationships between scheduling and interference behavior. The student will develop and evaluate prediction models that can distinguish true causal effects from spurious correlations, leading to more robust and interpretable interference prediction under dynamic network of conditions.

Possible Research Tasks:

• Review causal inference methods for wireless communication and time-series prediction. 
• Model interference in a 3D wireless network scenario with terrestrial and non-terrestrial nodes. 
• Identify relevant causal variables affecting interference.
• Develop causal interference prediction models. 
• Compare the proposed approach with conventional machine learning-based prediction methods.
• Evaluate prediction accuracy, robustness, and interpretability under different network conditions.

If you are interested in working on this project, please send your CV/resume and an up-to-date transcript of records from your Master's program.