DocumentCode
64449
Title
The Approximate Maximum-Likelihood Certificate
Author
Goldenberg, Idan ; Burshtein, David
Author_Institution
Sch. of Electr. Eng., Tel-Aviv Univ., Tel-Aviv, Israel
Volume
59
Issue
9
fYear
2013
fDate
Sept. 2013
Firstpage
6049
Lastpage
6059
Abstract
The confidence in the reliability of a codeword output by some (not necessarily optimal) decoding algorithm is discussed. A new property which relies on the linear programming (LP) decoder, the approximate maximum-likelihood certificate (AMLC), is introduced to address this issue as follows. First, the channel output vector is decoded by some symmetric decoder D, e.g., belief propagation or min-sum algorithm decoding. Second, the channel output vector is decoded by LP decoding. Third, if the decoding result of D is a codeword, its LP value is compared to the LP value of the LP decoding result (the latter need not be a codeword). If these two values are close, the AMLC holds. Using upper bounding techniques, we show that the conditional frame error probability given that the AMLC holds, is with some degree of confidence below a threshold. In channels with low noise, this threshold is orders of magnitude lower than the simulated frame error rate, and our bound holds with a very high degree of confidence. This is in stark contrast with standard Monte Carlo simulation, which would require excessively long runs to demonstrate like performance. When the AMLC holds, our approach thus provides the decoder with extra error detection capability, which is especially important in applications requiring high data integrity.
Keywords
Monte Carlo methods; approximation theory; channel coding; linear programming; maximum likelihood decoding; AMLC; LP decoder; Monte Carlo simulation; approximate maximum likelihood certificate; belief propagation; channel output vector; codeword output; data integrity; decoding algorithm; frame error probability; frame error rate simulation; linear programming; minsum algorithm decoding; symmetric decoder; Error probability; Maximum likelihood decoding; Monte Carlo methods; Parity check codes; Upper bound; Vectors; Linear programming (LP) decoding; low-density parity-check (LDPC) codes; maximum likelihood (ML) decoding; minimum distance; upper bounds on error probability;
fLanguage
English
Journal_Title
Information Theory, IEEE Transactions on
Publisher
ieee
ISSN
0018-9448
Type
jour
DOI
10.1109/TIT.2013.2263462
Filename
6516912
Link To Document