Fast envelope correlation for passive ranging.
Mika, Frank J.
Hippenstiel, Ralph D.
Titus, Harold A.
MetadataShow full item record
Application of classic triangulation methods will allow the location of a radar to be determined by passive sensors. Through the use of modern digital signal processing techniques this estimate can be made in a simpler fashion using a conventional receiver. In this thesis a technique is developed for time difference of arrival (TDOA) estimation using a frequency domain based correlation detector driven by an envelope detector. Time lag boundaries are defined on the output of the correlator. A fixed detection threshold is calculated to permit constant false alarm rate (CFAR) detection. The performance of the correlation detector is plotted as a receiver operating characteristic (ROC) curve as a function of signal to noise ratio (SNR) . An interactive MATLAB software program is provided to perform either spectral domain or time domain based correlation. Spectral domain based correlation uses the Fast Fourier Transform (FFT) . Implicit with the use of the FFT are finite arithmetic internal processing errors which are modeled as independent uncorrelated noise sources. A method is presented to account for SNR degradation at the output of the FFT.
Approved for public release; distribution is unlimited
Showing items related by title, author, creator and subject.
Smith, Peter E. (Monterey, California. Naval Postgraduate School, 1998-03-01);Energy spreading loss (ESL) is the reduction of the transmitted pulse energy level by spreading of the pulse in time due to multipath propagation. This energy spreading will reduce the effectiveness of mid-frequency tactical ...
VerPlanck, James Frederick (Monterey, California. Naval Postgraduate School, 1975-12);United States Coast Guard archival data covering two years was analyzed to determine if public relations efforts were affecting Coast Guard organization output. Quantitative measures of organization output and public ...
Kobayashi, Izumi (Monterey, California. Naval Postgraduate School, 2002-09);We propose two methods of constructing ensembles of classifiers. One method directly injects randomness into classification tree algorithms by choosing a split randomly at each node with probabilities proportional to the ...