Passive UAV Tracking in Wireless Networks

Abstract

Techniques for high accuracy tracking of the 3-D velocity, position, and cellular carrier frequency bias of unmanned aerial vehicles (UAVs) that communicate using a cellular network are presented. The tracking algorithm does <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">only</i> rely on asynchronous measurements of Doppler shift in a minimum of four base station sites. This number can be reduced to three, by combination with a round-trip time measurement from one site. The tracking is therefore independent of any capability of the UAV transceiver, like the global positioning system that may be disabled in the UAV by the pilot. Another significant advantage is that high precision time synchronization between base stations is not needed. The algorithm exploits interacting multiple model filtering, with the nonlinear measurement equations handled by extended Kalman filters. The low computational complexity, low sampling rate, and the uplink Doppler shift measurement resources enable joint tracking of several UAVs per cell. This makes the tracking technique relevant for supervision of UAV traffic using cellular connectivity, for example for border surveillance and surveillance of sensitive areas like airports where illegal flying may be extremely dangerous and costly. Theoretically, an observability analysis treats measurement configurations based on different combinations of Doppler shift and round-trip time. Simulations are used to illustrate the performance and to assess the initialization properties.