• DocumentCode
    3294032
  • Title

    Simulation study of variable step-size LMS algorithm in frequency domain for de-noising from MRS signal

  • Author

    Baofeng, Tian ; Chuandong, Jiang ; Huicui, Hao ; Qingming, Duan ; Jun, Lin

  • Author_Institution
    Coll. of Instrum. & Electr. Eng., Jilin Univ., Changchun, China
  • fYear
    2011
  • fDate
    15-17 April 2011
  • Firstpage
    4472
  • Lastpage
    4475
  • Abstract
    Magnetic Resonance Sounding (MRS) equipment for groundwater exploration has high receiving sensitivity, and it is vulnerably affected by high-level power-line harmonic when worked near the city and village. In order to effectively extract the weak signal, variable step-size LMS algorithm in frequency domain is studied to eliminate noise based on the principle of adaptive noise cancellation. Simulation results show that the algorithm can increase the signal to noise ratio more than 10dB, and the fitting error of parameters is about 10% in the condition that the frequency offset between signals and power-line harmonic is greater than 7Hz and with different SNR and water content. Therefor the variable step-size LMS algorithm in frequency domain can obtained better de-noising effect and parameter extraction accuracy.
  • Keywords
    adaptive signal processing; geophysical signal processing; groundwater; least mean squares methods; magnetic resonance; signal denoising; adaptive noise cancellation; frequency domain; groundwater exploration; magnetic resonance sounding signal de-noising; variable step-size LMS algorithm; water content; Fitting; Frequency domain analysis; Harmonic analysis; Instruments; Least squares approximation; Nuclear magnetic resonance; Power harmonic filters; LMS algorithm; MRS signal; adaptive noise cancellation; frequency domain; variable step size;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Electric Information and Control Engineering (ICEICE), 2011 International Conference on
  • Conference_Location
    Wuhan
  • Print_ISBN
    978-1-4244-8036-4
  • Type

    conf

  • DOI
    10.1109/ICEICE.2011.5778358
  • Filename
    5778358