Use of Particle Filtering in RSSI-Based Localization by Drone Base Stations

Abstract

Drone Base Stations (DBSs) provide flexible deployment and line-of-sight coverage opportunities, which led to many use cases, such as broadband Internet, military, surveillance, agriculture etc. DBSs can optimize and adapt their positions based on user location information. Especially in GPS-denied tactical scenarios ground user location estimation is an important problem. In this work we investigate particle filter as a method of user position estimation. We utilize the recently proposed air-to-ground pathloss model for RSSI-based location estimation. We investigate different DBS trajectories and various resampling methods. Finally, we show by simulations that particle filtering performs comparably to maximum likelihood estimation, which makes it a suitable alternative for localization and tracking.