Title :
Spectral graphs for quasi-cyclic LDPC codes
Author :
Tanner, R. Michael
Author_Institution :
Dept. of Comput. Sci., California Univ., Santa Cruz, CA, USA
Abstract :
Quasi-cyclic codes are described in terms of code equations on the spectral components of constituent cyclic codes. These define a constraint graph in the spectral domain. Here spectral graphs define low density parity check codes (LDPC) codes for which minimum distance can be bounded with algebraic and graph-based arguments. Examples include [42,22,8] and [155,64,20] regular LPDC codes
Keywords :
binary codes; cyclic codes; error detection codes; graph theory; spectral analysis; Tanner graphs; algebraic-based arguments; binary quasi-cyclic code; code equations; constraint graph; graph-based arguments; low density parity check codes; minimum distance; quasi-cyclic LDPC codes; regular LPDC codes; spectral components; spectral domain; spectral graphs; Computer science; Constraint theory; Equations; Frequency; Iterative algorithms; Iterative decoding; Matrix decomposition; Parity check codes; Sparse matrices; Testing;
Conference_Titel :
Information Theory, 2001. Proceedings. 2001 IEEE International Symposium on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-7123-2
DOI :
10.1109/ISIT.2001.936089
