Spatial Channel State Information Prediction with Generative AI: Towards Holographic Communication and Digital Radio Twin

TL;DR

The paper addresses real-time spatial-CSI prediction to enable holographic communication and digital radio twin capabilities for 6G. It proposes a digital radio twin framework and investigates neural radio tracing with a six-view image-to-image UNet to identify and reconstruct propagation paths from environmental data. Using a synthetic WiSegRT indoor dataset, the approach demonstrates learned propagation patterns and path alignment with ground truth, enabling 3D path reconstruction from multi-view inputs. The work highlights key challenges in accurate 3D environmental reconstruction, cross-environment generalization, and integration of sensing with planning for deterministic radio control in future wireless systems.

Abstract

As 5G technology becomes increasingly established, the anticipation for 6G is growing, which promises to deliver faster and more reliable wireless connections via cutting-edge radio technologies. However, efficient management method of the large-scale antenna arrays deployed by those radio technologies is crucial. Traditional management methods are mainly reactive, usually based on feedback from users to adapt to the dynamic wireless channel. However, a more promising approach lies in the prediction of spatial channel state information (spatial-CSI), which is an all-inclusive channel characterization and consists of all the feasible line-of-sight (LoS) and non-line-of-sight (NLoS) paths between the transmitter (Tx) and receiver (Rx), with the three-dimension (3D) trajectory, attenuation, phase shift, delay, and polarization of each path. Advances in hardware and neural networks make it possible to predict such spatial-CSI using precise environmental information, and further look into the possibility of holographic communication, which implies complete control over every aspect of the radio waves emitted. Based on the integration of holographic communication and digital twin, we proposed a new framework, digital radio twin, which takes advantages from both the digital world and deterministic control over radio waves, supporting a wide range of high-level applications. As a preliminary attempt towards this visionary direction, in this paper, we explore the use of generative artificial intelligence (AI) to pinpoint the valid paths in a given environment, demonstrating promising results, and highlighting the potential of this approach in driving forward the evolution of 6G wireless communication technologies.

Figures (6)

• Figure 1: Radio technologies that exploit spatial multiplexing: SU-MIMO, mMIMO, and RIS.
• Figure 2: Digital radio twin operating in a local intelligent base station
• Figure 3: Neural radio tracing to enable direct spatial-CSI prediction from images: A 3D input is projected to six views, then padded and inputted into a trained U-Net network, which predicts the 3D ray paths from three views. This network is able to implicitly learn radio propagation and interactions with objects.
• Figure 4: Training and validation loss of the proposed model for spatial-CSI prediction
• Figure 5: Sample results of the predicted ray paths. Top: predicted results; Bottom: Ground-truth from ray tracing simulator. Background is voided for better illustration. Although blurred and missing some parts, the predicted rays are well-aligned with true spatial paths. Both training and validation data are restricted to two ray interactions.