Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

JaGuard: Jamming Correction of GNSS Deviation with Deep Temporal Graphs

Ivana Kesić, Aljaž Blatnik, Carolina Fortuna, Blaž Bertalanič

TL;DR

JaGuard reframes GNSS jamming mitigation as a dynamic temporal graph regression problem using a receiver-centric star-graph representation and a single-layer HeteroGCLSTM to estimate 2D position deviations. The method delivers sub-10 cm MAE across diverse jamming types and power levels, with strong data efficiency and cross-device generalization, outperforming strong time-series baselines. Across controlled and mixed datasets, JaGuard demonstrates robust, real-time-friendly error correction that enhances navigation resilience without relying on raw measurements. This approach highlights the value of spatiotemporal graph modeling for reliable GNSS operation in interference-rich environments.

Abstract

Global Navigation Satellite Systems (GNSS) are increasingly exposed to intentional jamming, threatening reliability when accurate positioning and timing are most critical. We address this problem by formulating interference mitigation as a dynamic graph regression task and propose JaGuard, a receiver-centric temporal graph neural network that estimates and corrects latitude and longitude errors. At each 1 Hz epoch, the satellite-receiver scene is represented as a heterogeneous star graph with time-varying satellite attributes such as SNR, azimuth and elevation. A single-layer HeteroGCLSTM fuses one-hop spatial context with short-term temporal dynamics to produce a 2D deviation estimate. We evaluate JaGuard on data collected from two commercial receivers under controlled conducted jamming using three jammer types (CW, 3xCW, FM) and six power levels from -45 to -70 dBm, each repeated 50 times across pre-jam, jam, and recovery phases. JaGuard outperforms strong multivariate baselines (TSMixer, uniform CNN, Seq2Point) in all conditions. Under severe jamming at -45 dBm, it achieves 3.64-7.74 cm MAE, improving to 1.59-1.90 cm for -60 to -70 dBm. On mixed-mode datasets, it attains 3.78 cm MAE on GP01 and 4.25 cm on U-blox 10. With only 10 percent of the training data, JaGuard remains ahead, reaching about 20 cm MAE compared to 36-42 cm for the baselines.

JaGuard: Jamming Correction of GNSS Deviation with Deep Temporal Graphs

TL;DR

JaGuard reframes GNSS jamming mitigation as a dynamic temporal graph regression problem using a receiver-centric star-graph representation and a single-layer HeteroGCLSTM to estimate 2D position deviations. The method delivers sub-10 cm MAE across diverse jamming types and power levels, with strong data efficiency and cross-device generalization, outperforming strong time-series baselines. Across controlled and mixed datasets, JaGuard demonstrates robust, real-time-friendly error correction that enhances navigation resilience without relying on raw measurements. This approach highlights the value of spatiotemporal graph modeling for reliable GNSS operation in interference-rich environments.

Abstract

Global Navigation Satellite Systems (GNSS) are increasingly exposed to intentional jamming, threatening reliability when accurate positioning and timing are most critical. We address this problem by formulating interference mitigation as a dynamic graph regression task and propose JaGuard, a receiver-centric temporal graph neural network that estimates and corrects latitude and longitude errors. At each 1 Hz epoch, the satellite-receiver scene is represented as a heterogeneous star graph with time-varying satellite attributes such as SNR, azimuth and elevation. A single-layer HeteroGCLSTM fuses one-hop spatial context with short-term temporal dynamics to produce a 2D deviation estimate. We evaluate JaGuard on data collected from two commercial receivers under controlled conducted jamming using three jammer types (CW, 3xCW, FM) and six power levels from -45 to -70 dBm, each repeated 50 times across pre-jam, jam, and recovery phases. JaGuard outperforms strong multivariate baselines (TSMixer, uniform CNN, Seq2Point) in all conditions. Under severe jamming at -45 dBm, it achieves 3.64-7.74 cm MAE, improving to 1.59-1.90 cm for -60 to -70 dBm. On mixed-mode datasets, it attains 3.78 cm MAE on GP01 and 4.25 cm on U-blox 10. With only 10 percent of the training data, JaGuard remains ahead, reaching about 20 cm MAE compared to 36-42 cm for the baselines.

Paper Structure

This paper contains 18 sections, 5 equations, 7 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Related work
  3. Problem formulation
  4. Proposed method
  5. Proposed model
  6. Methodology and experimental details
  7. Dataset
  8. Model training and evaluation
  9. Comparison to the baseline models
  10. Comparative Analysis of Training Sample Ratios
  11. Ablation study
  12. Results
  13. Overall performance analysis
  14. Mixed dataset
  15. Analysis of Training Sample Ratios on Model Performance
  16. ...and 3 more sections

Figures (7)

  • Figure 1: L1-band jamming profiles overlaid on the GNSS spectrum.
  • Figure 2: From physical constellation to graph snapshot
  • Figure 3: Temporal graph sequence
  • Figure 4: model architecture
  • Figure 5: Positional deviation of the GP01 GNSS receiver across 50 repeated measurements under $3\times$CW jamming.
  • ...and 2 more figures