CSIT-Free Model Aggregation for Federated Edge Learning via Reconfigurable Intelligent Surface

We study over-the-air model aggregation in federated edge learning (FEEL) systems, where channel state information at the transmitters (CSIT) is assumed to be unavailable. leverage reconfigurable intelligent surface (RIS) technology align cascaded coefficients for xmlns:xlink="http://www.w3.org/1999/xlink">CSIT-free aggregation. To this end, we jointly optimize RIS and receiver by minimizing error under alignment constraint. then develop a difference-of-convex algorithm resulting non-convex optimization. Numerical experiments on image classification show that proposed method able achieve similar accuracy as state-of-the-art CSIT-based solution, demonstrating efficiency of our approach combating lack CSIT.