Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Needle in a Haystack: Tracking UAVs from Massive Noise in Real-World 5G-A Base Station Data

Chengzhen Meng, Chenming He, Yidong Jiang, Xiaoran Fan, Dequan Wang, Lingyu Wang, Jianmin Ji, Yanyong Zhang

Abstract

The potential usage of UAVs in daily life has made monitoring them essential. However, existing systems for monitoring UAVs typically rely on cameras, LiDARs, or radars, whose limited sensing range or high deployment cost hinder large-scale adoption. In response, we develop BSense, the first system that tracks UAVs by leveraging point clouds from commercial 5G-A base stations. The key challenge lies in the dominant number of noise points that closely resemble true UAV points, resulting in a noise-to-UAV ratio over 100:1. Therefore, identifying UAVs from the raw point clouds is like finding a needle in a haystack. To overcome this, we propose a layered framework that filters noise at the point, object, and trajectory levels. At the raw point level, we observe that noise points from different spatial regions exhibit distinguishable and consistent signal fingerprints, which we can model to identify and remove them. At the object level, we design spatial and velocity consistency checks to identify false objects, and further compute confidence scores by aggregating these checks over multiple frames for more reliable discrimination. At the final trajectory level, we propose a Transformer-based network that captures multi-frame motion patterns to filter the few remaining false trajectories. We evaluated BSense on a commercial 5G-A base station deployed in an urban environment. The UAV was instructed to fly along 25 distinct trajectories across 54 cases over 7 days, yielding 155 minutes of data with more than 14,000 frames. On this dataset, our system reduces the number of false detections from an average of 168.05 per frame to 0.04, achieving an average F1 score of 95.56% and a mean localization error of 4.9 m at ranges up to 1,000 m.

Needle in a Haystack: Tracking UAVs from Massive Noise in Real-World 5G-A Base Station Data

Abstract

The potential usage of UAVs in daily life has made monitoring them essential. However, existing systems for monitoring UAVs typically rely on cameras, LiDARs, or radars, whose limited sensing range or high deployment cost hinder large-scale adoption. In response, we develop BSense, the first system that tracks UAVs by leveraging point clouds from commercial 5G-A base stations. The key challenge lies in the dominant number of noise points that closely resemble true UAV points, resulting in a noise-to-UAV ratio over 100:1. Therefore, identifying UAVs from the raw point clouds is like finding a needle in a haystack. To overcome this, we propose a layered framework that filters noise at the point, object, and trajectory levels. At the raw point level, we observe that noise points from different spatial regions exhibit distinguishable and consistent signal fingerprints, which we can model to identify and remove them. At the object level, we design spatial and velocity consistency checks to identify false objects, and further compute confidence scores by aggregating these checks over multiple frames for more reliable discrimination. At the final trajectory level, we propose a Transformer-based network that captures multi-frame motion patterns to filter the few remaining false trajectories. We evaluated BSense on a commercial 5G-A base station deployed in an urban environment. The UAV was instructed to fly along 25 distinct trajectories across 54 cases over 7 days, yielding 155 minutes of data with more than 14,000 frames. On this dataset, our system reduces the number of false detections from an average of 168.05 per frame to 0.04, achieving an average F1 score of 95.56% and a mean localization error of 4.9 m at ranges up to 1,000 m.

Paper Structure

This paper contains 27 sections, 20 equations, 26 figures.

Table of Contents

  1. Introduction
  2. Background
  3. BSense Design
  4. System Overview
  5. Noise Point Filtering through Noise Fingerprint Modeling
  6. Analysis of Noise Fingerprint
  7. Multivariate Gaussian Modeling.
  8. Similarity-based Noise Filtering.
  9. Automatic Model Parameter Updating.
  10. False Object Filtering Using Spatial- and Velocity-based Confidences
  11. Point Clustering.
  12. Spatial-based Confidence.
  13. Velocity-based Confidence.
  14. Confidence-based Object Discrimination.
  15. Object Tracking and False Trajectory Filtering
  16. ...and 12 more sections

Figures (26)

  • Figure 1: A single frame of real-world data. The number of noise points exceeds UAV points by a ratio of 174:1.
  • Figure 2: The value ranges of point features for UAV and noise exhibit substantial overlap.
  • Figure 3: Persistent noise points give rise to massive long-term false trajectories.
  • Figure 4: BSense adopts a layered framework consisting of (1) noise point filtering, (2) false object filtering, and (3) false trajectory filtering, which progressively suppress false detections to yield clean UAV trajectories.
  • Figure 5: Noise point filtering through fingerprint modeling. The 3D space is partitioned into cubes, and the noise within each cube is modeled using a Gaussian distribution. Points are then filtered based on their similarity to the corresponding distribution.
  • ...and 21 more figures