![[2509.19405] Improving Outdoor Multi-cell Fingerprinting-based Positioning via Mobile Data Augmentation](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2509.19405] Improving Outdoor Multi-cell Fingerprinting-based Positioning via Mobile Data Augmentation
Summary
This paper presents a novel mobile data augmentation framework to enhance outdoor multi-cell fingerprinting-based positioning, improving accuracy by up to 30%.
Why It Matters
Accurate outdoor positioning is crucial for various applications in cellular networks, yet traditional methods are often limited by data sparsity and high costs. This research offers a practical solution that leverages existing mobile data to improve positioning services, making it relevant for network operators aiming to enhance user experience and operational efficiency.
Key Takeaways
- Introduces a lightweight framework for mobile data augmentation.
- Utilizes kernel density estimation and k-nearest neighbors for improved positioning.
- Achieves up to a 30% reduction in median positioning error in complex areas.
- Framework is training-free and interpretable, supporting privacy-aware workflows.
- Demonstrates effectiveness using real-world data from an Italian mobile operator.
Computer Science > Networking and Internet Architecture arXiv:2509.19405 (cs) [Submitted on 23 Sep 2025 (v1), last revised 23 Feb 2026 (this version, v2)] Title:Improving Outdoor Multi-cell Fingerprinting-based Positioning via Mobile Data Augmentation Authors:Tony Chahoud, Lorenzo Mario Amorosa, Riccardo Marini, Luca De Nardis View a PDF of the paper titled Improving Outdoor Multi-cell Fingerprinting-based Positioning via Mobile Data Augmentation, by Tony Chahoud and 3 other authors View PDF HTML (experimental) Abstract:Accurate outdoor positioning in cellular networks is hindered by sparse, heterogeneous measurement collections and the high cost of exhaustive site surveys. This paper introduces a lightweight, modular mobile data augmentation framework designed to enhance multi-cell fingerprinting-based positioning using operator-collected minimization of drive test (MDT) records. The proposed approach decouples spatial and radio-feature synthesis: kernel density estimation (KDE) models the empirical spatial distribution to generate geographically coherent synthetic locations, while a k-nearest-neighbor (KNN)-based block produces augmented per-cell radio fingerprints. The architecture is intentionally training-free, interpretable, and suitable for distributed or on-premise operator deployments, supporting privacy-aware workflows. We both validate each augmentation module independently and assess its end-to-end impact on fingerprinting-based positioning using a real-world M...
