Anteumbler: Non-Invasive Antenna Orientation Error Measurement for WiFi APs

TL;DR

Anteumbler tackles the critical problem of WiFi localization errors caused by AP antenna orientation misalignment. It introduces a non-invasive, CSI-based spatial-angle model that estimates each antenna's 3-D orientation by exploiting the peak received power, coupled with an iterative, plane-based search and a plane-intersection geometry to remove distance effects. The approach achieves median elevation and azimuth errors below $8^{\circ}$ across diverse hardware, layouts, and environments, and demonstrably improves reverse localization and user localization accuracy in real deployments. This work enables accurate, long-term antenna orientation calibration without AP hardware or firmware changes, improving the robustness of WiFi sensing systems in real-world deployments.

Abstract

The performance of WiFi-based localization systems is affected by the spatial accuracy of WiFi AP. Compared with the imprecision of AP location and antenna separation, the imprecision of AP's or antenna's orientation is more important in real scenarios, including AP rotation and antenna irregular tilt. In this paper, we propose Anteumbler that non-invasively, accurately and efficiently measures the orientation of each antenna in physical space. Based on the fact that the received power is maximized when a Tx-Rx antenna pair is perfectly aligned, we construct a spatial angle model that can obtain the antennas' orientations without prior knowledge. However, the sampling points of traversing the spatial angle need to cover the entire space. We use the orthogonality of antenna directivity and polarization and adopt an iterative algorithm to reduce the sampling points by hundreds of times, which greatly improves the efficiency. To achieve the required antenna orientation accuracy, we eliminate the influence of propagation distance using a dual plane intersection model and filter out ambient noise. Our real-world experiments with six antenna types, two antenna layouts and two antenna separations show that Anteumbler achieves median errors below 6 degree for both elevation and azimuth angles, and is robust to NLoS and dynamic environments. Last but not least, for the reverse localization system, we deploy Anteumbler over LocAP and reduce the antenna separation error by 10 mm, while for the user localization system, we deploy Anteumbler over SpotFi and reduce the user localization error by more than 1 m.

Figures (12)

• Figure 1: Motivation: (a) Four orientation errors of AP or antennas (yaw, roll, pitch, irregular tilt). (b) The antennas have different tilted angles in real scenarios. (c) The localization error vs. the orientation error of AP or antenna.
• Figure 2: Antenna radiation or reception: (a) Antenna system for Tx-Rx pair. (b) Radiation pattern of omnidirectional antenna. (c) Effect of directivity on received power. (d) Effect of polarization on received power.
• Figure 3: System overview: platform, data processing, estimating the orientations of target antennas.
• Figure 4: Unknown antenna parameters effect removing: (a) For a target-local antenna pair, the received power can be expressed as the function of $\phi$ and $\theta$, and reaches the maximum when ($\phi=0,\theta=0$). (b) Through coordinate transformation, we convert the unknown ($\phi,\theta$) of the antenna pair to the known ($\alpha,\beta$) of the local antenna; note that ($\alpha=\alpha_{obj}, \beta=\beta_{obj}$) corresponds to ($\phi=0,\theta=0$). (c) For a target antenna with fixed orientation, the received power can be expressed as the function of $\alpha$ and $\beta$ of the local antenna, and reaches the maximum when ($\alpha=\alpha_{obj},\beta=\beta_{obj}$).
• Figure 5: Time cost optimization and propagation distance influence removing: (a) The local antenna exhibits the same received power at different ($\alpha,\beta$). (b) Using the orthogonality of $\phi$ and $\theta$, we convert $\alpha$ and $\beta$ to $\phi$ and $\theta$ respectively, and remove the influence of propagation distance based on the geometric principle of plane intersection. (c) The orientation of the target antenna can be quickly obtained by solving for the relative maximum of each vertical plane and then for their absolute maximum.