Near-field positioning improves multi-user localization with confidence
Reliable Near-Field Multi-User Positioning Informed by Two-Stage MUSIC
Artificial IntelligenceInformation Theory
Summary
Finding an exact location of multiple users nearby using wireless signals is tricky because signals bounce around, causing confusion. The authors developed a new method called MUSIC-Net that uses a smart combination of existing signal processing with deep learning to pinpoint where multiple users are. Their system also gives a reliable estimate of how confident it is in these location guesses. Tests show this approach is more accurate and trustworthy than earlier methods, especially in complicated signal environments.
What this means in practice
- •For wireless network engineers: Improve location-based services in complex environments with guaranteed confidence bounds on user positions.
- •For indoor navigation system developers: Enable accurate and reliable positioning of multiple users indoors despite signal reflections and blockages.
Authors
Jiaying Li, Haifeng Wen, Changsheng You, Yuanwei Liu, Hong Xing
Abstract
Near-field localization is a promising technique for high-resolution multi-user positioning in future wireless systems, but its performance is often degraded by scattering-induced coherent propagation. Existing near-field localization methods, which require separate parameter estimation and path/source association, suffer from high computation overhead and accumulated errors, and usually do not provide any guarantee on reliability. In this paper, we propose \emph{MUSIC-Net}, an end-to-end near-field positioning deep learning (DL) framework informed by two-stage MUltiple SIgnal Classification (MUSIC) in mixed line-of-sight (LoS) and non-LoS (NLoS) multi-path scenarios, which embeds the two-stage MUSIC objects into training to isolate the LoS-related signal subspace and to identify a surrogate distance. The proposed framework directly recovers multi-user positions without the need for involved NLoS parameter estimation or path/source association. Furthermore, we introduce split conformal prediction (SCP) to move beyond point-estimation-based positioning towards statistically guaranteed (confidence) set estimation for all users. Numerical results show that the proposed MUSIC-Net achieves lower mean positioning error (MPER) than existing benchmarks and yields tighter SCP-calibrated prediction regions, demonstrating both accurate LoS localization and efficient uncertainty quantification (UQ) in coherent multi-path environments.