KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Search Results

Now showing 1 - 3 of 3
  • Some of the metrics are blocked by your 
    Item type:Item,
    A simple design for large girth LDPC matrix
    (2017-07-01)
    Choomchuay, Somsak
    This paper describes an alternative method for designing regular quasi-cyclic LDPC codes with large girths. The permutation matrix was designed based on the geometric sequence and prime number field. The design starts with a portion of the full-range matrix with column weight of 3. With a decided code rate, elements in the rows are extended by adding a certain column of the same full-range matrix into the designing matrix. Resulted codes can offer relatively good performance compared to the published literatures.
  • Some of the metrics are blocked by your 
    Item type:Item,
    Construction of quasi-cyclic LDPC codes form SFT structure and cyclic shift
    (2011-12-01)
    Timakul, Sekson
    ;
    Choomchuay, Somsak
    This paper presents an alternative a method for designing a parity matrix of low density parity check codes. The proposed method can be viewed as a generalized version of Sridara-Fuja-Tanner (SFT) technique which is ideal for regular quasi cyclic code design. Upon the current investigation, the designed code has delivered a comparable BER-SNR performance when compared with that of more complicate designed matrix. The designed code offers similar performance when compared with Sridara-Fuja-Tanner code and Quadratic Congruences structure. In additions, the resulted matrix also has a desirable structure that it suits well the hardware implementation. © 2011 IEEE.
  • Some of the metrics are blocked by your 
    Item type:Item,
    A design of nonprime block irregular LDPC codes via CRT
    (2010-12-01)
    Chusin, Chalit
    ;
    Prasartkaew, Chutima
    ;
    Timakul, Sekson
    ;
    Choomchuay, Somsak
    This paper presents a new design of irregular LDPC codes that supports arbitrary block length. We propose the efficient construction method when nonprime size sub-matrices are used. The problem where GCD(L<inf>1</inf>,L <inf>2</inf>) that left unsolved has been tackled. We also consider the case GCD(L<inf>1</inf>,L<inf>2</inf>)=1 but L<inf>1</inf> or L<inf>2</inf> is a nonprime number. The results show that our designed codes have superior performance compared to MAC. The resulted codes still hold the attractive properties of LDPC codes. The performances of interleaved codes are higher than noninterleaved codes and it goes higher as SNR is increased. ©2010 IEEE.