• DocumentCode
    1525715
  • Title

    Radar target discrimination using a hybrid E-pulse fourth-order moments technique

  • Author

    Gallego, A. ; Ruiz, D.P. ; Carrión, M.C.

  • Author_Institution
    Dept. de Fisica Aplicada, Granada Univ., Spain
  • Volume
    144
  • Issue
    6
  • fYear
    1997
  • fDate
    12/1/1997 12:00:00 AM
  • Firstpage
    355
  • Lastpage
    360
  • Abstract
    A variant of the original extinction pulse (E-pulse) technique for radar target discrimination is proposed. The scheme includes, before convolution with the E-pulse discriminant waveform, a new stage in which the fourth-order moments sequence of the unknown-target impulse-response, r(t), is estimated. To justify this inclusion, it has been shown previously that, by using a particular set of one-dimensional slices on the fourth-order moments space and an appropriately defined `covariance-type´ estimator, this moments sequence retains the same damped complex exponential structure as signal r(t), thus only containing the natural resonances of the unknown target. Simulation results clearly show that significant improvements are achieved by using the fourth-order moments sequence of r(t) as the unknown-target signature, instead of the original signal r(t) in the E-pulse discrimination scheme. These results are especially notable when high levels of additive Gaussian noise, white or coloured, are present in r(t)
  • Keywords
    Gaussian noise; convolution; higher order statistics; radar detection; radar signal processing; radar target recognition; white noise; E-pulse discriminant waveform; additive Gaussian noise; coloured noise; convolution; covariance-type estimator; damped complex exponential structure; extinction pulse; fourth-order moments sequence; hybrid E-pulse fourth-order moments; natural resonances; radar target discrimination; simulation results; target impulse-response; unknown-target signature; white noise;
  • fLanguage
    English
  • Journal_Title
    Radar, Sonar and Navigation, IEE Proceedings -
  • Publisher
    iet
  • ISSN
    1350-2395
  • Type

    jour

  • DOI
    10.1049/ip-rsn:19971498
  • Filename
    646938