Application of narrow-band and broadband signal processing techniques to the acoustic detection and localization of aircraft

Abstract

The acoustic spectrum of a transiting aircraft is shown to be ‘‘time varying’’ when received by a stationary observer. This phenomenon is an example of the Doppler effect, which is exploited to localize aircraft having narrow-band lines in their source spectra; for example, propeller-driven aircraft and helicopters. For the present case of a turboprop aircraft, the application of time-varing spectral analysis techniques to the acoustic data from a single sensor enables the instantaneous propeller blade rate to be estimated at short-time intervals during the aircraft transit. The temporal variation of the Doppler blade rate is required as an input to the localization algorithm, which provides estimates of the aircraft’s speed and range at the closest point of approach, as well as the source (or rest) frequency of the propeller blade rate. Next, a broadband cross-correlation technique is applied to the acoustic data from a pair of spatially separated microphones to demonstate the automatic detection of an aircraft transit. Short integration times and Doppler compensation are required to track the transit of a fast aircraft.