Neural Positioning Without External Reference

2025-11-20

AI Summary

This work addresses the challenge of high-precision user equipment (UE) localization in the absence of external reference positioning systems. We propose the first fully self-supervised neural localization framework, which trains end-to-end solely on relative displacement commands executed by a commercial robotic platform and externally measured Wi-Fi/5G NR channel state information (CSI)—requiring neither ground-truth position labels nor specialized hardware. By formulating self-supervised signals grounded in relative motion constraints, our method achieves near-state-of-the-art localization accuracy across diverse real-world scenarios, ranging from small-scale line-of-sight to large-scale non-line-of-sight environments. Compared to conventional approaches reliant on costly external trackers or labor-intensive precise annotations, our framework significantly reduces deployment cost and calibration complexity, thereby enhancing scalability and practicality for large-scale wireless localization systems.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Machine Learning: Unsupervised & Self-Supervised Learning
• Computer Vision: Motion & Tracking

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Channel state information (CSI)-based user equipment (UE) positioning with neural networks -- referred to as neural positioning -- is a promising approach for accurate off-device UE localization. Most existing methods train their neural networks with ground-truth position labels obtained from external reference positioning systems, which requires costly hardware and renders label acquisition difficult in large areas. In this work, we propose a novel neural positioning pipeline that avoids the need for any external reference positioning system. Our approach trains the positioning network only using CSI acquired off-device and relative displacement commands executed on commercial off-the-shelf (COTS) robot platforms, such as robotic vacuum cleaners -- such an approach enables inexpensive training of accurate neural positioning functions over large areas. We evaluate our method in three real-world scenarios, ranging from small line-of-sight (LoS) areas to larger non-line-of-sight (NLoS) environments, using CSI measurements acquired in IEEE 802.11 Wi-Fi and 5G New Radio (NR) systems. Our experiments demonstrate that the proposed neural positioning pipeline achieves UE localization accuracies close to state-of-the-art methods that require externally acquired high-precision ground-truth position labels for training.

Problem

• Eliminates need for external reference positioning systems in neural localization
• Trains positioning networks using CSI and robot displacement commands only
• Enables inexpensive training of accurate positioning over large areas

Innovation

• Uses CSI and robot displacement for training
• Avoids external reference positioning systems
• Works with Wi-Fi and 5G NR systems