3D angle-of-arrival positioning using von Mises-Fisher distribution

TL;DR

This paper introduces a von Mises-Fisher distribution-based model for 3D angle-of-arrival positioning, improving accuracy and stability over traditional methods especially near poles by representing measurements as unit vectors.

Contribution

It proposes a novel VMF distribution approach for AOA measurements, reducing nonlinearity and discontinuities inherent in azimuth-elevation models.

Findings

VMF-based filters outperform normal distribution methods in accuracy

Improved handling of measurements near poles

Reduces nonlinearity in AOA positioning

Abstract

We propose modeling an angle-of-arrival (AOA) positioning measurement as a von Mises-Fisher (VMF) distributed unit vector instead of the conventional normally distributed azimuth and elevation measurements. Describing the 2-dimensional AOA measurement with three numbers removes discontinuities and reduces nonlinearity at the poles of the azimuth-elevation coordinate system. Our computer simulations show that the proposed VMF measurement noise model based filters outperform the normal distribution based algorithms in accuracy in a scenario where close-to-pole measurements occur frequently.