Fast Chase algorithm with an application in turbo decoding

Turbo product codes (TPCs) provide an attractive alternative to recursive systematic convolutional (RSC)-based turbo systems. Rather than employ trellis-based decoders, an algebraic decoder may be repeatedly employed in a low-complexity, soft-input/soft-output errors-and-erasures decoder such as the...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on communications Vol. 49; no. 10; pp. 1693 - 1699
Main Authors: Hirst, S.A., Honary, B., Markarian, G.
Format: Journal Article
Language:English
Published: New York IEEE 01.10.2001
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
Algebra
Algorithms
Block codes
Computation
Convolutional codes
Decoders
Decoding
Degradation
Error correction codes
Maximum likelihood decoding
Maximum likelihood estimation
Product codes
Recursive
Redundancy
Testing
Turbo codes
ISSN:0090-6778, 1558-0857
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Turbo product codes (TPCs) provide an attractive alternative to recursive systematic convolutional (RSC)-based turbo systems. Rather than employ trellis-based decoders, an algebraic decoder may be repeatedly employed in a low-complexity, soft-input/soft-output errors-and-erasures decoder such as the Chase algorithm. Taking motivation from efficient forced erasure decoders, this implementation re-orders the Chase algorithm's repeated decodings such that the inherent computational redundancy is greatly reduced without degrading performance. The result is a highly efficient fast Chase implementation. The algorithm presented here is principally applicable to single error-correcting codes although consideration is also given to the more general case. The new decoder's value in practical turbo schemes is demonstrated via application to decoding of the (64,57,4) extended Hamming TPC.
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
content type line 23
ISSN:0090-6778
1558-0857
DOI:10.1109/26.957387