A harmonic spectral beamformer for the enhanced localization of propeller-driven aircraft

Abstract

The following paper presents several array processing techniques that may be used to enhance the localization of acoustic source targets, such as UAVs. A review of common methods is first provided, followed by several algorithms developed to reduce computational loads for the application of concern. A beamforming method is proposed that exploits the properties of harmonic narrowband signals, such as that generated by propeller-driven aircraft to enhance direction of arrival accuracy. In addition, a regional contraction search algorithm is proposed to minimize computational loads associated with the presented localization technique. A brief comparison of the proposed method to that of standard frequency domain beamformers is also provided using both theoretical analysis and experimental data. For the case of target localization between two moving fixed-wing UAVs, it was found that the proposed harmonic spectral beamforming method increased localization accuracy by 50% over the standard steered response power approach.