A Survey on Federated Learning in RIS, STAR-RIS, and BD-RIS-Assisted Wireless Networks

Abstract

The advent of sixth-generation (6G) wireless networks has spurred growing interest in technologies capable of simultaneously meeting stringent requirements for capacity, latency, energy efficiency (EE), and security. Among these, Reconfigurable Intelligent Surfaces (RIS), Simultaneously Transmitting and Reflecting RIS (STAR-RIS), and Beyond-Diagonal RIS (BD-RIS) have emerged as transformative enablers for next-generation communication systems. This article presents an extensive survey on the integration of Federated Learning (FL) within RIS, STAR-RIS, and BD-RIS-assisted wireless networks, highlighting their complementary potential in creating intelligent, adaptive, and secure communication environments. We provide a comprehensive analysis of recent research efforts exploring the synergy between FL and these advanced reconfigurable surfaces, spanning applications such as the Internet of Things (IoT), vehicular networks, edge computing, and secure communications. The survey introduces a detailed taxonomy encompassing network architectures, optimization objectives, resource allocation mechanisms, and algorithmic frameworks. Moreover, we classify existing works according to five key dimensions: RIS architecture, FL mechanism, optimization goal, optimization constraints, and application environment, thereby offering a clearer perspective on how each contribution fits within the broader literature and enabling direct comparisons among different approaches. In addition, we also examine the unique challenges associated with integrating FL in such systems, including accurate channel estimation for hybrid reflecting/transmitting surfaces, dynamic configuration of RIS/STAR-RIS/BD-RIS elements, communication-computation trade-offs in distributed learning, and resilience to adversarial threats. Finally, we identify key open research problems and outline future directions, such as leveraging artificial intelligence (AI) for unified optimization across heterogeneous reconfigurable surfaces, adaptive orchestration of resources in highly dynamic environments, and the establishment of standardized performance metrics. By consolidating the state-of-the-art, this survey serves as a foundational reference for researchers and practitioners aiming to design efficient, robust, and intelligent wireless networks empowered by FL and advanced reconfigurable surfaces.