A measurement method to discriminate aircraft fly-over noise

Currently aircraft noise monitoring systems use a mesh of single microphones distributed around an airport to continuously sample the noise level. This fact requires a manual process of aircraft noise event detection and classification in order to distinguish aircraft events from the rest of noise events in the recording. In the present paper a 3-meter-long 12-microphone linear array is used to automatically obtain a background noise free aircraft noise recording. The beamforming process separates the noise impinging in the array from above (potential aircraft noise) and the noise impinging from below (urban noise and reflections), the results are enhanced by the use of a trigger condition on the difference between both. The theoretical results reveals that the background noise in the aircraft noise recording can be attenuated by about 8 dB if the microphone array is optimally placed. The experimental tests shows that even in non optimal placements the array still provides better results than a single microphone if the threshold value in the trigger condition is properly set.