Scalable multipath detection from standard GNSS correlators using multifrequency diversity

Abstract

Multipath propagation fundamentally limits wireless positioning in dense urban environments, yet existing mitigation methods often rely on specialized hardware or poorly interpretable data-driven models. We propose a multi-frequency cross-product area (MF-CPA) framework for multipath detection using only standard early-prompt-late I/Q correlator outputs, requiring no receiver modification. By analyzing correlator sensitivity to relative delay, phase, Doppler, and attenuation, we show that the six-branch I/Q structure enables a physically interpretable detection metric. Multi-frequency fusion further mitigates frequency-dependent blind zones inherent to single-frequency methods. Hardware-in-the-loop simulations in a 3D urban canyon demonstrate detection rates of 50–80% at low false-discovery levels across GPS, BDS, and Galileo signals. Real-world experiments confirm consistent spatial detection patterns, though validation remains qualitative due to the lack of ground-truth channel-state labels. The proposed framework provides a scalable and interpretable solution for robust multipath monitoring in next-generation wireless navigation systems.