Suboptimal Decoding of Linear Codes
Suboptimal decoding algorithms of linear codes in an arbitrary “symmetric” memoryless channel are considered. The decoding error probability є is upper bounded by twice the error probability є e of maximum likelihood (ML) decoding. For the q-ary codes of length n → ∞ and code rate R the asymptotic equality є ~ є e holds, while the number of decoding operations is upper bounded by the value qn(c+0(1)), where 0(1) → 0 and c = min (R(1-R), (1-R)/2). For channels with discrete (quantized) output the better estimate c = R(1-R)/(1+R) is obtained. Suboptimal coverings with polynomial construction complexity are also considered.
KeywordsError Probability Linear Code Code Rate Cyclic Code Light Vector
Unable to display preview. Download preview PDF.
- L. Levitin and C. Hartmann, A new approach to the general minimum distance decoding problem: The zero-neighbours algorithm IEEE Trans. Inform. Theory, vol. IT-31, pp. 379 - 384, 1985.Google Scholar
- J.T. Coffey and R.M.F. Goodman, The complexity of information set decoding. IEEE Trans. Inform. Theory, vol. IT-36, pp. 1031 - 1037, 1990.Google Scholar
- I.I. Dumer, On minimum distance decoding of linear codes. Int. Proc. of the Fifth Intern. Workshop on Information Theory. Convolutional Codes; Multiuser Communication. Moscow, pp. 50 - 52, Jan. 1991.Google Scholar
- J.K. Wolf, Efficient maximum likelihood decoding of linear codes using a trellis. IEEE Trans. Inform. Theory, vol. IT-24, pp. 76 - 80, 1978.Google Scholar
- P. Erdos and J. Spencer, Probabilistic methods in combinatorics. Akademiai Kiado, Budapest, 1974.Google Scholar