Robots improve radio transmitter location and direction in cluttered rooms
MAGNETAR: Multipath-Guided Spatial Posteriors for Transmitter Pose Inference in the Upper Mid-Band
Robotics
Summary
Finding where a hidden radio device is can be tricky because signals can bounce off walls and furniture, creating confusing clues. The authors designed a system called MAGNETAR that looks at these bounced signals and guesses both the transmitter's exact spot and the direction it's facing. MAGNETAR uses smart neural networks trained with computer simulations and some real-world data to get better results. Compared to older methods, it gives more detailed and accurate guesses when working in complex indoor spaces with directional antennas.
What this means in practice
- •For robotics engineers: Improve indoor robot localization by estimating both position and heading of radio transmitters from single RF signal snapshots in cluttered environments.
- •For wireless network planners: Enhance placement and troubleshooting of RF devices by more accurately inferring transmitter poses in complex indoor settings using multipath signal information.
Authors
Haozhe Lei, Ruibin Chen, Yuhan Jiang, Ali Rasteh, Aditya Dhananjay, Sundeep Rangan
Abstract
Robots that localize a radio transmitter need more than a point estimate: in cluttered rooms, one measurement is often consistent with several transmitter locations and, because upper-mid-band antennas are directional, several headings. We present MAGNETAR, which infers a joint posterior over planar transmitter position and heading from a single asynchronous radio-frequency (RF) multipath snapshot, represented by angle-of-arrival and signal-to-noise-ratio estimates, given the room layout and receiver pose. Among our five neural scorers, MAGNETAR adopts a shared 2D U-Net conditioned on each candidate heading, jointly normalizing scores over a discretized position-heading grid. Training uses real-to-sim-calibrated 10 GHz simulations and a small measured subset. Grid-based joint posteriors outperform parametric ones on held-out simulations, the heading-conditioned scorer transfers best to robotic measurements, and fusing joint posteriors improves on fusing position-only marginals.