• DocumentCode
    1796406
  • Title

    A non-linear LLR approximation for LDPC decoding over impulsive noise channels

  • Author

    Yi Hou ; Rongke Liu ; Ling Zhao

  • Author_Institution
    Sch. of Electron. & Inf. Eng., Beihang Univ., Beijing, China
  • fYear
    2014
  • fDate
    13-15 Oct. 2014
  • Firstpage
    86
  • Lastpage
    90
  • Abstract
    The performance of Low-Density Parity-Check codes over symmetric alpha-stable impulsive noise channels is considered in this paper. Due to the lack of closed-form expressions for the probability density functions of almost all the alpha-stable distributions, an efficient non-linear approximation method for channel log-likelihood ratio is presented to achieve high decoding performance covering the range of characteristic exponent from 1 to 2 and avoid the use of complicated numerical integrations. The simulations for both long and short codes show that the parameter selection of the proposed method is not sensitive to the noise level. Even fixed parameter setting for the proposed non-linear approximation can obtain a satisfying decoding performance.
  • Keywords
    approximation theory; channel coding; decoding; impulse noise; integration; parity check codes; probability; LDPC decoding; alpha-stable distributions; channel log-likelihood ratio; characteristic exponent; closed-form expressions; complicated numerical integrations; fixed parameter setting; low-density parity-check codes; nonlinear LLR approximation method; parameter selection; probability density functions; symmetric alpha-stable impulsive noise channels; Bit error rate; Decoding; Linear approximation; Noise; Numerical models; Parity check codes; LDPC codes; impulsive noise; non-linear approximation; symmetric alpha-stable distribution;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Communications in China (ICCC), 2014 IEEE/CIC International Conference on
  • Conference_Location
    Shanghai
  • Type

    conf

  • DOI
    10.1109/ICCChina.2014.7008248
  • Filename
    7008248