6G communications using deep learning

a) modal-domain channel tracking and prediction using score-guided diffusion model and pretrained LLM for holographic MIMO systems (HMIMO)

- Modal-domain based hybrid beamforming and beamfocusing for near- and far-field users

b) Sensing: Localization using modal-domain method

c) practical reconfigurable intelligent surface (RIS)

- modal-domain channel tracking and prediction using score-guided diffusion model and pretrained LLM for RIS-assisted MIMO and HMIMO systems

- modal-domain hybrid beamforming and beamfocusing

- wideband RIS

d) low latency and low energy consumption virtual function network design using service chain graphs (related to above work on GNN and GCNN) using reinforcement learning and LLM

e) model-agnostic federated learning

I would prefer the intern to stay for 3 months, with background in wireless communications and deep learning, including generative AI models, deep reinforcement learning, optimization and a strong foundational mathematical background. I can offer NT$12,000/month for 3 months, with the remaining $24,000 to offset the cost of plane ticket and/or accommodation.