Initial Value Selection Algorithm for User Position Detection Method Based on Sequential Computation Using Unmanned Aerial Vehicles

Unmanned aircraft systems (UASs) have been developed and studied as temporal communication systems for emergency and rescue services during disasters. In a typical UAS model, several unmanned aerial vehicles (UAVs) are used to provide services over a large area. The UAV is comprised of a transmitter and receiver to communicate with terrestrial stations and terminals. Therefore, the carrier frequencies experience Doppler shifts due to the variations in the line-of-sight velocity between the UAV and the terrestrial terminal. Thus, by observing multiple Doppler shifts from different UAVs, it is possible to detect the position of a user based on the sequential computation. In this paper, we newly proposed a selection algorithm of the initial value for the least squares method in position detection procedure based on 1) linear approximation method of hyperbolic line formed by Doppler shifts, 2) decision process using relationships between velocity vector and its normal vector of UAV and plus or minus value of Doppler shift, and 3) repeatedly conducting the initial value detection procedure. This paper reports that the computer simulation assuming two UAVs shows the high performance of initial value correctly selected rate and provides good performance equivalent to that of ideal case in terms of position detection accuracy.