Deep learning (DL) techniques have demonstrated strong performance in compressing and reconstructing channel state information (CSI) while reducing feedback overhead in massive MIMO systems. A key challenge, however, is their reliance on extensive site-specific training data, whose real-world collection incurs significant overhead and limits scalability across deployment sites. To address this, we propose leveraging site-specific digital twins to assist the training of DL-based CSI compression models. The digital twin integrates an electromagnetic (EM) 3D model of the environment, a hardware model, and ray tracing to produce site-specific synthetic CSI data, allowing DL models to be trained without the need for extensive real-world measurements. We further develop a fidelity analysis framework that decomposes digital twin quality into four key aspects: 3D geometry, material properties, ray tracing, and hardware modeling. We explore how these factors influence the reliability of the data and model performance. To enhance the adaptability to real-world environments, we propose a refinement strategy that incorporates a limited amount of real-world data to fine-tune the DL model pre-trained on the digital twin dataset. Evaluation results show that models trained on site-specific digital twins outperform those trained on generic datasets, with the proposed refinement method effectively enhancing performance using limited real-world data. The simulations also highlight the importance of digital twin fidelity, especially in 3D geometry, ray tracing, and hardware modeling, for improving CSI reconstruction quality. This analysis framework offers valuable insights into the critical fidelity aspects, and facilitates more efficient digital twin development and deployment strategies for various wireless communication tasks.

Deep learning (DL) approaches have demonstrated high performance in compressing and reconstructing the channel state information (CSI) and reducing the CSI feedback overhead in massive MIMO systems. One key challenge, however, with the DL approaches is the demand for extensive training data. Collecting this real-world CSI data incurs significant overhead that hinders the DL approaches from scaling to a large number of communication sites. To address this challenge, we propose a novel direction that utilizes site-specific digital twins to aid the training of DL models. The proposed digital twin approach generates site-specific synthetic CSI data from the EM 3D model and ray tracing, which can then be used to train the DL model without real-world data collection. To further improve the performance, we adopt online data selection to refine the DL model training with a small real-world CSI dataset. Results show that a DL model trained solely on the digital twin data can achieve high performance when tested in a real-world deployment. Further, leveraging domain adaptation techniques, the proposed approach requires orders of magnitude less real-world data to approach the same performance of the model trained completely on a real-world CSI dataset.

Reconfigurable intelligent surfaces (RISs) have attracted increasing interest due to their ability to improve the coverage, reliability, and energy efficiency of millimeter wave (mmWave) communication systems. However, designing the RIS beamforming typically requires large channel estimation or beam training overhead, which degrades the efficiency of these systems. In this paper, we propose to equip the RIS surfaces with visual sensors (cameras) that obtain sensing information about the surroundings and user/basestation locations, guide the RIS beam selection, and reduce the beam training overhead. We develop a machine learning (ML) framework that leverages this visual sensing information to efficiently select the optimal RIS reflection beams that reflect the signals between the basestation and mobile users. To evaluate the developed approach, we build a high-fidelity synthetic dataset that comprises co-existing wireless and visual data. Based on this dataset, the results show that the proposed vision-aided machine learning solution can accurately predict the RIS beams and achieve near-optimal achievable rate while significantly reducing the beam training overhead.

Can reconfigurable intelligent surfaces (RISs) operate in a standalone mode that is completely transparent to the 3GPP 5G initial access process? Realizing that may greatly simplify the deployment and operation of these surfaces and reduce the infrastructure control overhead. This paper investigates the feasibility of building standalone/transparent RIS systems and shows that one key challenge lies in determining the user equipment (UE)-side RIS beam reflection direction. To address this challenge, we propose to equip the RISs with multi-modal sensing capabilities (e.g., using wireless and visual sensors) that enable them to develop some perception of the surrounding environment and the mobile users. Based on that, we develop a machine learning framework that leverages the wireless and visual sensors at the RIS to select the high-performance beams between the base station (BS) and UEs and enable standalone/transparent RIS operation for 5G high-frequency systems. Using a high-fidelity synthetic dataset with co-existing wireless and visual data, we extensively evaluate the performance of the proposed framework. Experimental results demonstrate that the proposed approach can accurately predict the BS and UE-side candidate beams, and that the standalone RIS beam selection solution is capable of realizing near-optimal achievable rates with significantly reduced beam training overhead.