Blind Quadratic and Time Frequency Based Detectors from Training Data

dc.citation.bibtexNameinproceedingsen_US
dc.citation.conferenceNameIEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)en_US
dc.contributor.authorJones, Douglas L.en_US
dc.contributor.authorSayeed, Akbar M.en_US
dc.contributor.orgDigital Signal Processing (http://dsp.rice.edu/)en_US
dc.date.accessioned2007-10-31T00:48:18Zen_US
dc.date.available2007-10-31T00:48:18Zen_US
dc.date.issued1995-01-20en_US
dc.date.modified2004-01-22en_US
dc.date.note2004-01-09en_US
dc.date.submitted1995-01-20en_US
dc.descriptionConference paperen_US
dc.description.abstractTime-frequency based methods, particularly quadratic (Cohen's-class) representations, are often considered for detection in applications ranging from sonar to machine monitoring. We propose a method of obtaining near-optimal quadratic detectors directly from training data using Fisher's optimal linear discriminant to design a quadratic detector. This detector is optimal in terms of Fisher's scatter criterion as applied to the quadratic outer product of the data vector, and in early simulations appears to closely approximate the true optimal quadratic detector. By relating this quadratic detector to an equivalent operation on the Wigner distribution of a signal, we derive near-optimal time-frequency detectors. A simple example demonstrates the excellent performance of the method.en_US
dc.identifier.citationD. L. Jones and A. M. Sayeed, "Blind Quadratic and Time Frequency Based Detectors from Training Data," 1995.en_US
dc.identifier.doihttp://dx.doi.org/10.1109/ICASSP.1995.480410en_US
dc.identifier.urihttps://hdl.handle.net/1911/19984en_US
dc.language.isoengen_US
dc.subjectquadratic representationsen_US
dc.subjectFisher's optimal linear discriminanten_US
dc.subject.keywordquadratic representationsen_US
dc.subject.keywordFisher's optimal linear discriminanten_US
dc.subject.otherTime Frequency and Spectral Analysisen_US
dc.titleBlind Quadratic and Time Frequency Based Detectors from Training Dataen_US
dc.typeConference paperen_US
dc.type.dcmiTexten_US
Files
Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
Jon1995Jan5BlindQuad.PS
Size:
384.29 KB
Format:
Postscript Files