Acoustic Localization Using an L-Shape Microphone Array for Multirotor Drones

Coherent signal subspace methods are a common means of dealing with the localization of broadband sources. The noise of a drone is a typical broadband signal, and its spectrum has obvious acoustic characteristics. To improve the positioning accuracy of drones in a low signal-to-noise environment, this paper proposes a processing method based on harmonic focusing by using the harmonic regularity information of the drone acoustic signal. The signal-to-noise ratio is indirectly enhanced by artificially eliminating the interference information outside the harmonic frequencies. Experiments show that the method is more resistant to interference and improves the positioning accuracy compared to the narrow-band positioning method and a conventional focusing method.