Browse Prior Art Database

Acoustic Broadband Signal Detection With DIFAR Buoys

IP.com Disclosure Number: IPCOM000035974D
Original Publication Date: 1989-Aug-01
Included in the Prior Art Database: 2005-Jan-28
Document File: 3 page(s) / 61K

Publishing Venue

IBM

Related People

Galkowski, PJ: AUTHOR [+2]

Abstract

Detection and bearing estimation of submerged vehicles through analysis of their broadband signals that may be 20 dB or lower than background noise is possible by measuring differential time of arrival (T) and differential time rate (T) or Doppler shift of the target signal received at a pair of DIFAR sonobuoys.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 53% of the total text.

Page 1 of 3

Acoustic Broadband Signal Detection With DIFAR Buoys

Detection and bearing estimation of submerged vehicles through analysis of their broadband signals that may be 20 dB or lower than background noise is possible by measuring differential time of arrival (T) and differential time rate (T) or Doppler shift of the target signal received at a pair of DIFAR sonobuoys.

Broadband signals are obtained from two sonobouys placed such that the second buoy lies anywhere on 100-600-foot. radius from the first. Each buoy produces three signals, omnidirectional (O), northerly (N) and easterly (E), and for analysis four signals are required, such as 01, 02, N2 and E2.

(Image Omitted)

In Fig. 1a, these signals are each coupled to a respective bandpass filter 1 to restrict signal and noise spectrums and avoid an alias problem. These analog signals are coupled in turn to respective sample-and-hold circuits 2, and the outputs are converted to digital values stored in buffer 3. Data are read out from buffers 3 first-in, first-out. A data block is formed of n samples with n typically 1024, padded with up to 1024 zeros and windowed via Hanning or other windows. These modified data blocks are coupled into digital Fourier transform signal processors 4. The complex signal from channel 01 is directed to spectral compensation block 5. Assuming the signal has a differential time rate, T, the spectral compensation block computes the spectrum for the signal with no T. (Intially, T control is set to zero and no change to input spectra SO1(K) is made, so that the output signal is identical to the input signal at T=0.) The doppler- compensated signal SO1(K) is conjugated and cross power spectra are gener ated for the other channels by multiplication with the conjugate of SO1(K) at circuits 6. The resultant complex power spectra are stored in accumulators 7 in Fig. 1b.

Control signal m specifies the number of cross power spectra summed and controls signal integration time where total effective integration time is mnt/2, with n being the number of samples used in forming the data block input to the fourier transform processors 4 and t the period of the sampling clock. Once m samples of cross power spectra are accumulated, each 50% overlapped in time, the accumulated cross power spectra are filtered to the desired bandwidth by zeroing complex numbers for frequency components outside the band of interest. Retained spectral components are whitened at 8; that is, each complex component of the accumulated cross power spectra is normalized to unity magnitude. Such processed cross power spectral signals are then coupled to an inverse digital Fourier transform processor 9 that generates the correlation function at T and for time lag values spa...