The modified generalized cross-correlation method is used for finding the time difference of arrival (TDOA) and resolving the angle of arrival toward moving targets in real-time. It addresses the decorrelation due to differential Doppler effect leading to relative time companding (RTC) between sensors. The proposed method efficiently approximates the RTC prior to the computation of cross-correlation, thus working faster than methods using maximization of the passive ambiguity function and other approximation methods. The algorithm is based on computing the spectrum of deskewed short-time cross-correlator (DSTC) in the frequency domain by averaging a short-term spectrogram computed on overlapped segments of analysis in a time window. The same spectrogram is used to compute magnitude-squared coherence (MSC), then by maximizing the average MSC in the band of interest, the RTC is found. It is used to compute a single inverse Fourier transform. This method is applied to data collected by the Acoustic Aircraft Detection system that was developed at Stevens and that comprises of a volumetric 5-microphone cluster connected to a data acquisition and an embedded computer that processes the data in real-time. Examples demonstrate the improved TDOA finding for aircraft within an airport at a larger distance by using longer analysis time window and maintenance of tracking at close range, thanks to the RTC compensation. [This work was supported by DHS’s S&T Directorate.]