• DocumentCode
    284716
  • Title

    Detection of transients using discrete wavelet transform

  • Author

    Petropulu, Athina P.

  • Author_Institution
    Dept. of Electr. & Comput. Eng., Northeastern Univ., Boston, MA, USA
  • Volume
    2
  • fYear
    1992
  • fDate
    23-26 Mar 1992
  • Firstpage
    477
  • Abstract
    A scheme for the detection of transient signals with unknown waveforms and arrival times is presented. The detection is performed in the discrete wavelet transform (DWT) domain, where the presence of a transient is indicated by a peak at a location that depends on the values of the dilation and translation parameters. By choosing these parameters appropriately one can control the sharpness of this peak as well as the distance between different peaks that correspond to different arrival times. As a result, transients that partially overlap in time can be resolved in the DWT domain. The DWT domain preserves the arrival time information even when the transient is corrupted by additive zero-mean noise at a very low signal-to-noise ratio
  • Keywords
    signal detection; transients; wavelet transforms; SNR; additive zero-mean noise; arrival times; dilation parameters; discrete wavelet transform; signal-to-noise ratio; transient signal detection; translation parameters; Additive noise; Discrete wavelet transforms; Gaussian noise; Gaussian processes; Signal representations; Signal resolution; Signal to noise ratio; Statistics; Wavelet domain;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Acoustics, Speech, and Signal Processing, 1992. ICASSP-92., 1992 IEEE International Conference on
  • Conference_Location
    San Francisco, CA
  • ISSN
    1520-6149
  • Print_ISBN
    0-7803-0532-9
  • Type

    conf

  • DOI
    10.1109/ICASSP.1992.226016
  • Filename
    226016