Feature Learning for Neural-Network-Based Positioning with Channel State Information

TL;DR

This paper introduces a deep learning-based CSI positioning system that learns features directly from raw measurements and fuses temporal data, significantly improving accuracy over existing methods.

Contribution

It proposes a novel CSI-based positioning pipeline that learns features from raw data and fuses temporal information, enhancing accuracy in real-world indoor scenarios.

Findings

Reduces mean distance error by up to 2.5 times

Demonstrates effectiveness with real-world LoS and non-LoS data

Improves accuracy over state-of-the-art methods

Abstract

Recent channel state information (CSI)-based positioning pipelines rely on deep neural networks (DNNs) in order to learn a mapping from estimated CSI to position. Since real-world communication transceivers suffer from hardware impairments, CSI-based positioning systems typically rely on features that are designed by hand. In this paper, we propose a CSI-based positioning pipeline that directly takes raw CSI measurements and learns features using a structured DNN in order to generate probability maps describing the likelihood of the transmitter being at pre-defined grid points. To further improve the positioning accuracy of moving user equipments, we propose to fuse a time-series of learned CSI features or a time-series of probability maps. To demonstrate the efficacy of our methods, we perform experiments with real-world indoor line-of-sight (LoS) and non-LoS channel measurements. We show that CSI feature learning and time-series fusion can reduce the mean distance error by up to 2.5$\boldsymbol\times$ compared to the state-of-the-art.