Title :
On hybrid stack decoding algorithms for block codes
Author :
Hguado, L.E. ; Farrell, P.G.
Author_Institution :
Commun. Res. Group, Manchester Univ., UK
fDate :
1/1/1998 12:00:00 AM
Abstract :
This correspondence presents sequential algorithms for soft decision decoding of linear block codes. They use a stack algorithm based on the trellis of the code. We are interested in the trellis as a means to avoid path-decoding repetitions. As well, the possibility of bidirectional decoding offers a chance to increase the likelihood of explored paths. We have developed three successive algorithms that offer a good decrement in the overall complexity, and mainly in the most complex decoding case, while giving near-maximum-likelihood performance. This is important since it determines the maximum buffer size necessary in the decoder
Keywords :
block codes; computational complexity; linear codes; maximum likelihood decoding; sequential decoding; trellis codes; bidirectional decoding; complexity; hybrid stack decoding algorithms; linear block codes; maximum buffer size; near-maximum-likelihood performance; sequential algorithms; soft decision decoding; trellis code; Block codes; Error correction codes; Hamming weight; Information science; Information theory; Iterative algorithms; Iterative decoding; Maximum likelihood decoding; Sufficient conditions; Welding;
Journal_Title :
Information Theory, IEEE Transactions on