deposit_your_work

Blind Quadratic and Time Frequency Based Detectors from Training Data

Files in this item

Files Size Format View
Jon1995Jan5BlindQuad.PS 393.5Kb application/postscript View/Open

Show full item record

Item Metadata

Title: Blind Quadratic and Time Frequency Based Detectors from Training Data
Author: Jones, Douglas L.; Sayeed, Akbar M.
Type: Conference Paper
Keywords: quadratic representations; Fisher's optimal linear discriminant
Citation: D. L. Jones and A. M. Sayeed,"Blind Quadratic and Time Frequency Based Detectors from Training Data," in IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP),
Abstract: Time-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.
Date Published: 1995-01-20

This item appears in the following Collection(s)

  • ECE Publications [1030 items]
    Publications by Rice University Electrical and Computer Engineering faculty and graduate students
  • DSP Publications [508 items]
    Publications by Rice Faculty and graduate students in digital signal processing.