Bayesian Jammer Localization with a Hybrid CNN and Path-Loss Mixture of Experts

2025-10-23

AI Summary

To address inaccurate RSS field reconstruction and degraded localization accuracy in urban environments caused by GNSS multipath and signal blockage, this paper proposes a physics-informed Bayesian mixture-of-experts model. The method integrates log-linear pooling CNNs with an analytical path-loss model to enhance spatial contextual modeling; employs Laplace approximation for joint uncertainty quantification of position and RSS fields; and explicitly incorporates building height maps to encode urban propagation geometry. Experiments on ray-tracing data demonstrate substantial improvements in interference source localization accuracy. Crucially, the estimated uncertainty heatmaps precisely concentrate over urban canyons and interference source regions—thereby achieving high accuracy, strong interpretability, and reliable uncertainty estimation.

Technology Category

• Intelligent Robots: State Estimation
• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• 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

Abstract

Global Navigation Satellite System (GNSS) signals are vulnerable to jamming, particularly in urban areas where multipath and shadowing distort received power. Previous data-driven approaches achieved reasonable localization but poorly reconstructed the received signal strength (RSS) field due to limited spatial context. We propose a hybrid Bayesian mixture-of-experts framework that fuses a physical path-loss (PL) model and a convolutional neural network (CNN) through log-linear pooling. The PL expert ensures physical consistency, while the CNN leverages building-height maps to capture urban propagation effects. Bayesian inference with Laplace approximation provides posterior uncertainty over both the jammer position and RSS field. Experiments on urban ray-tracing data show that localization accuracy improves and uncertainty decreases with more training points, while uncertainty concentrates near the jammer and along urban canyons where propagation is most sensitive.

Problem

• Locates GNSS jammers in urban areas with multipath interference
• Reconstructs signal strength field using hybrid physical-neural network model
• Quantifies uncertainty in jammer position and signal propagation

Innovation

• Hybrid Bayesian mixture-of-experts framework fuses physical and neural models
• Log-linear pooling combines path-loss model with CNN using building maps
• Bayesian inference with Laplace approximation provides posterior uncertainty estimates