Multi-View Wireless Sensing via Conditional Generative Learning: Framework and Model Design

2025-05-19

AI Summary

This work addresses high-precision wireless sensing in multi-base-station–user-equipment cooperative scenarios, proposing a physics-informed conditional diffusion generative framework for joint 3D point-cloud and electromagnetic (EM) property reconstruction. Methodologically: (1) a physics-aware encoder adaptively fuses multi-view channel state information (CSI) while embedding EM propagation priors; (2) a two-stage architecture—supporting arbitrary numbers and layouts of base stations—is built upon spatial positional encoding and a weighted-loss conditional diffusion model. The key contribution lies in the first deep integration of an EM physical model into the diffusion process, enabling end-to-end disentanglement and reconstruction of multi-view CSI features. Experiments demonstrate substantial improvements over state-of-the-art methods in both geometric shape fidelity and EM parameter accuracy, while exhibiting strong generalization capability and deployment flexibility.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

In this paper, we incorporate physical knowledge into learning-based high-precision target sensing using the multi-view channel state information (CSI) between multiple base stations (BSs) and user equipment (UEs). Such kind of multi-view sensing problem can be naturally cast into a conditional generation framework. To this end, we design a bipartite neural network architecture, the first part of which uses an elaborately designed encoder to fuse the latent target features embedded in the multi-view CSI, and then the second uses them as conditioning inputs of a powerful generative model to guide the target's reconstruction. Specifically, the encoder is designed to capture the physical correlation between the CSI and the target, and also be adaptive to the numbers and positions of BS-UE pairs. Therein the view-specific nature of CSI is assimilated by introducing a spatial positional embedding scheme, which exploits the structure of electromagnetic(EM)-wave propagation channels. Finally, a conditional diffusion model with a weighted loss is employed to generate the target's point cloud from the fused features. Extensive numerical results demonstrate that the proposed generative multi-view (Gen-MV) sensing framework exhibits excellent flexibility and significant performance improvement on the reconstruction quality of target's shape and EM properties.

Problem

• High-precision target sensing using multi-view CSI
• Fusing latent target features via conditional generation
• Reconstructing target shape and EM properties adaptively

Innovation

• Multi-view CSI fused via bipartite neural network
• Spatial positional embedding for EM-wave correlation
• Conditional diffusion model with weighted loss