Reproduce “JEPA-Inspired Latent Field Prediction for Sparse-Pilot MIMO-OFDM Channel Reconstruction” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Yuzhi Yang, Brahim Mefgouda, Hang Zou, Lina Bariah, Anis Bara, Yuhuan Lu, Hao Zhang, Merouane Debbah
The field world model predicts latent channel features from historical observations and combines them with sparse, noisy current pilots to reconstruct channel state information in MIMO-OFDM systems. Inspired by JEPA, it uses predicted time-frequency tokens as a prior rather than replacing measurements. In Chicago ray-tracing simulations, covariance-based variants outperform pilot-only estimation and direct training, with evidence limited to that setting.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.