Spatial and Temporal Generalization of CSI-based Neural Positioning

2026-06-16

AI Summary

This study addresses the limited practical relevance of random data splitting in existing channel state information (CSI)-based neural localization methods, which fails to reflect real-world demands for spatiotemporal generalization. Focusing on standards-compliant Wi-Fi and 5G NR systems, the work systematically evaluates the generalization performance of neural localization models across unseen areas, trajectories, and measurements collected weeks apart in realistic indoor and outdoor environments. To this end, a lightweight Transformer architecture is proposed, achieving higher localization accuracy with fewer parameters compared to a multilayer perceptron (MLP). Experimental results demonstrate that both architectures exhibit strong spatiotemporal generalization capabilities, yet the proposed Transformer consistently outperforms the MLP in both accuracy and model efficiency.

Abstract

Channel state information (CSI)-based neural positioning learns a mapping from CSI measurements to user equipment (UE) positions using neural networks. However, most existing performance evaluations utilize randomly partitioned train/test CSI-dataset splits, which fail to reflect the generalization requirements of practical deployments and present optimistic results. In this paper, we study the spatial and temporal generalization of neural positioning with standard-compliant Wi-Fi and 5G NR systems for three real-world CSI datasets acquired in indoor and outdoor environments. We assess generalization with two different architectures, a conventional multilayer perceptron (MLP) and a novel transformer architecture, to unseen spatial regions, unseen UE trajectories, and CSI measurement campaigns separated by one week. Our experiments show that both architectures generalize well in space and time, and the proposed transformer consistently outperforms the MLP in positioning accuracy while requiring fewer model parameters.

Problem

• CSI-based neural positioning
• spatial generalization
• temporal generalization
• practical deployment
• performance evaluation

Innovation

• spatial generalization
• temporal generalization
• CSI-based neural positioning
• transformer architecture
• 5G NR