DocumentCode
3047767
Title
Trellis-canonical generator matrices for convolutional codes
Author
Lin, Wei ; Mceliece, Robert J. ; Xu, Meina
Author_Institution
Dept. of Electr. Eng., California Inst. of Technol., Pasadena, CA, USA
fYear
1997
fDate
29 Jun-4 Jul 1997
Firstpage
286
Abstract
It was asserted in McEliece without proof, that a canonical generator matrix G(D) is trellis-canonical if and only if G(D) has the property that the span-length of the corresponding scalar matrix “G¯” cannot be reduced by a row operation of the form Row[m]=Row[n]Ds+Row[m], where s is an integer in the range 0⩽s⩽L and m≠n. In this paper, we prove a stronger result, viz., a basic PGM is trellis-canonical if and only if it is “row-reduced”. An efficient algorithm for converting a basic PGM into a trellis-canonical PGM is presented. We also correct an error in the general algorithm given in Lin and McEliece (1995)
Keywords
convolutional codes; matrix algebra; trellis codes; convolutional codes; general algorithm; row operation; row-reduced PGM; scalar matrix; span-length; trellis-canonical PGM; trellis-canonical generator matrices; Bifurcation; Convolutional codes; Error correction;
fLanguage
English
Publisher
ieee
Conference_Titel
Information Theory. 1997. Proceedings., 1997 IEEE International Symposium on
Conference_Location
Ulm
Print_ISBN
0-7803-3956-8
Type
conf
DOI
10.1109/ISIT.1997.613207
Filename
613207
Link To Document