What I work on
My PhD work sits at the intersection of non-terrestrial networks and deep reinforcement learning: teaching UAVs and high-altitude platforms to position and move themselves to keep ground users connected. Around that core, I work more broadly across AI/ML and wireless communications.
Deep Reinforcement Learning
Policy-gradient and actor-critic methods for continuous positioning and trajectory control, applied to keep aerial base stations optimally placed under changing conditions.
- PPO
- TD3
- Trajectory optimization
- Positioning
- Autonomous decision-making
- Continuous control
- Multi-agent coordination
Non-Terrestrial Networks
Integrating UAVs and high-altitude platform stations (HAPS) with terrestrial networks to keep ground and aerial users connected — including a GPS-free framework, using only radio-sensing and angle-of-arrival measurements, for coordinating multiple UAV base stations in disaster-response and wind-disturbed maritime scenarios.
- UAV base stations
- HAPS
- Terrestrial/NTN integration
- Disaster response
- Maritime networks
AI & Machine Learning
Broader work in federated learning, autoencoder-based classification and detection, data valuation, and large language models.
- Federated Learning
- CNN Autoencoders
- Large Language Models
- Data valuation
Wireless Communications
MIMO systems and next-generation wireless standards, including indirect SINR estimation from SSB and CSI-RS reference signals in 5G NR.
- MIMO
- 5G/6G
- SINR estimation
- SSB / CSI-RS