• DocumentCode
    1768636
  • Title

    De-noising filter study of the sound-source using the DWT

  • Author

    Bo-Yeon Hwang ; Jong-Ho Han ; Min-Gyu Kim ; Jang-Myung Lee

  • Author_Institution
    Dept. of Electron., Electr. & Comput. Eng., Pusan Nat. Univ., Pusan, South Korea
  • fYear
    2014
  • fDate
    22-25 Oct. 2014
  • Firstpage
    856
  • Lastpage
    861
  • Abstract
    In this paper, the de-noising filter study of the sound-source using the DWT is introduced. We suggest the de-noising filter using the Discrete Wavelet Transform (DWT). The DWT has been performed to remove out the noises from the signals. In order to confirm the performance of suggested filtering algorithm, we estimate the distance and angle from the microphone to the sound-source and compare the filtered signal to the original signal. To calculate the distance and angle, we use the cross-correlation that is possible to estimate the time difference for sound waves reaching more than two sound-source obtaining devices. Also, through the sound-source obtaining devices and carrying out the geometric interpretation, the location of each sound-source can be found.
  • Keywords
    correlation methods; discrete wavelet transforms; filtering theory; geometry; microphone arrays; signal denoising; speech recognition; DWT; angle estimation; cross-correlation; discrete wavelet transform; distance estimation; filtering algorithm; geometric interpretation; human voice recognition; microphone array; sound waves; sound-source denoising filter study; sound-source obtaining devices; time difference estimation; Discrete wavelet transforms; Equations; Noise; Noise reduction; Writing; De-noising Filter; Discrete Wavelet Transform; Microphone array; Sound Source;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Control, Automation and Systems (ICCAS), 2014 14th International Conference on
  • Conference_Location
    Seoul
  • ISSN
    2093-7121
  • Print_ISBN
    978-8-9932-1506-9
  • Type

    conf

  • DOI
    10.1109/ICCAS.2014.6987900
  • Filename
    6987900