A Beamspace Multi-sources DOA Estimation Method for UAV Cluster Systems

The UAV (Unmanned-Aerial-Vehicle) cluster system has been widely used for multi-sources DOA estimation. To achieve better performance, the antenna array is equipped onboard, but the element number is quite restricted because of the finite volume of UAV unit. The UAV unit cooperation is taken to extend the equivalent aperture and spatial degrees of freedom, but it still suffers the restricted RF chain number. Besides, due to the imprecise locations and the asynchronous local oscillator problem, there is always a random phase difference between adjacent UAV units. All of these problems degrade the performance of multi-sources DOA estimation by UAV cluster system. In order to resolve these problems in practice, we propose a beamspace DOA Matrix method. It reduces the RF chain number by beamspace signal processing, and a rotation invariance between beam clusters is established to deal with the random phase difference. The simulation result shows that the proposed method can realize the multi-sources DOA estimation even with random phase difference. It achieves similar estimation performance with the traditional element space method by only 2dB degradation, and both the RF chain number and the computation burden are greatly reduced.