A Novel Bayesian Optimization Framework for RIS-Assisted Wireless Networks

Reconfigurable Intelligent Surface (RIS) emerges as a promising technology for shaping the wireless propagation environment. In this paper, we introduce a novel Bayesian optimization framework to address the challenge of joint channel estimation and optimization in complex wireless environments. To reduce the dimensionality of large-scale RIS, the framework incorporates a unique physics-informed structure embedding utilizing angular and spatial domain basis vectors. We validate our approach through laboratory experiments conducted with software-defined radio equipment, considering scenarios where the receivers are located in near-field or far-field regions. The results demonstrate superior channel gain per iteration compared to existing methods.