Attention-Based Fusion of IQ and FFT Spectrograms with AoA Features for GNSS Jammer Localization

TL;DR

This work tackles GNSS jammer localization under challenging multipath by introducing an attention-based fusion that jointly processes IQ samples, FFT spectrograms, and 22 AoA features to detect, classify, and localize jamming sources. The methodology combines three processing paths—vision-encoder-based spectrograms, time-series IQ models, and AoA feature digital processing—with a shared attention fusion and dense regression heads, evaluated against a large indoor moving-jammer dataset. Results show the fusion approach achieves superior localization accuracy over classical methods and several ML baselines, highlighting the benefit of multi-representation fusion and AoA-aware features for robust GNSS interference characterization. The dataset and findings advance practical GNSS situational awareness and inform counter-measure design in multipath-rich indoor environments.

Abstract

Jamming devices disrupt signals from the global navigation satellite system (GNSS) and pose a significant threat by compromising the reliability of accurate positioning. Consequently, the detection and localization of these interference signals are essential to achieve situational awareness, mitigating their impact, and implementing effective counter-measures. Classical Angle of Arrival (AoA) methods exhibit reduced accuracy in multipath environments due to signal reflections and scattering, leading to localization errors. Additionally, AoA-based techniques demand substantial computational resources for array signal processing. In this paper, we propose a novel approach for detecting and classifying interference while estimating the distance, azimuth, and elevation of jamming sources. Our benchmark study evaluates 128 vision encoder and time-series models to identify the highest-performing methods for each task. We introduce an attention-based fusion framework that integrates in-phase and quadrature (IQ) samples with Fast Fourier Transform (FFT)-computed spectrograms while incorporating 22 AoA features to enhance localization accuracy. Furthermore, we present a novel dataset of moving jamming devices recorded in an indoor environment with dynamic multipath conditions and demonstrate superior performance compared to state-of-the-art methods.

Figures (28)

• Figure 1: Overview of our pipeline. Raw measurements are collected from mobile jamming devices and processed into IQ samples. Vision encoders are trained with FFT-computed spectrograms, and combined with time-series models and statistical features. After concatenation, all features are passed through dense layers and ReLU activation for three distinct losses.
• Figure 2: Overview of the adapted fusion baseline McAFF zeng_gong_liu for GNSS jammer localization.
• Figure 3: Jammer.
• Figure 4: Wall 1.
• Figure 5: Wall 2.