Channel capacity for a given decoding metric

I. Csiszár, Prakash Narayan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

We address the rate of transmission attainable on a given channel when the decoding rule is specified, perhaps suboptimally. We concentrate on decoders, termed d-decoders, which accept the codeword x 'closest' to the received sequence y in the sense of a metric d(x,y), defined for sequences as an additive extension of a single-letter metric. We provide a simple sufficient condition for Cd(W) = C(W), and more generally, for the equality of the d-capacities for two different decoding metrics d and d̄. This is followed by a sufficient condition for Ceo (W) = C(W). We then show that the lower bound on d-capacity given previously by past studies, is not tight in general but Cd(W) > iff this bound is positive. The 'product space' improvement of the lower bound is considered, and a 'product space characterization' of Ceo(W) is obtained. We also determine the e.o. capacity of a deterministic arbitrarily varying channel defined by a bipartite graph, and show that it equals capacity.

Original languageEnglish
Title of host publicationIEEE International Symposium on Information Theory - Proceedings
PublisherIEEE
Publication statusPublished - 1994
EventProceedings of the 1994 IEEE International Symposium on Information Theory - Trodheim, Norw
Duration: Jun 27 1994Jul 1 1994

Other

OtherProceedings of the 1994 IEEE International Symposium on Information Theory
CityTrodheim, Norw
Period6/27/947/1/94

Fingerprint

Channel capacity
Decoding

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Csiszár, I., & Narayan, P. (1994). Channel capacity for a given decoding metric. In IEEE International Symposium on Information Theory - Proceedings IEEE.

Channel capacity for a given decoding metric. / Csiszár, I.; Narayan, Prakash.

IEEE International Symposium on Information Theory - Proceedings. IEEE, 1994.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Csiszár, I & Narayan, P 1994, Channel capacity for a given decoding metric. in IEEE International Symposium on Information Theory - Proceedings. IEEE, Proceedings of the 1994 IEEE International Symposium on Information Theory, Trodheim, Norw, 6/27/94.
Csiszár I, Narayan P. Channel capacity for a given decoding metric. In IEEE International Symposium on Information Theory - Proceedings. IEEE. 1994
Csiszár, I. ; Narayan, Prakash. / Channel capacity for a given decoding metric. IEEE International Symposium on Information Theory - Proceedings. IEEE, 1994.
@inproceedings{b39dd9266f114d0e90b2363eb030fc49,
title = "Channel capacity for a given decoding metric",
abstract = "We address the rate of transmission attainable on a given channel when the decoding rule is specified, perhaps suboptimally. We concentrate on decoders, termed d-decoders, which accept the codeword x 'closest' to the received sequence y in the sense of a metric d(x,y), defined for sequences as an additive extension of a single-letter metric. We provide a simple sufficient condition for Cd(W) = C(W), and more generally, for the equality of the d-capacities for two different decoding metrics d and d̄. This is followed by a sufficient condition for Ceo (W) = C(W). We then show that the lower bound on d-capacity given previously by past studies, is not tight in general but Cd(W) > iff this bound is positive. The 'product space' improvement of the lower bound is considered, and a 'product space characterization' of Ceo(W) is obtained. We also determine the e.o. capacity of a deterministic arbitrarily varying channel defined by a bipartite graph, and show that it equals capacity.",
author = "I. Csisz{\'a}r and Prakash Narayan",
year = "1994",
language = "English",
booktitle = "IEEE International Symposium on Information Theory - Proceedings",
publisher = "IEEE",

}

TY - GEN

T1 - Channel capacity for a given decoding metric

AU - Csiszár, I.

AU - Narayan, Prakash

PY - 1994

Y1 - 1994

N2 - We address the rate of transmission attainable on a given channel when the decoding rule is specified, perhaps suboptimally. We concentrate on decoders, termed d-decoders, which accept the codeword x 'closest' to the received sequence y in the sense of a metric d(x,y), defined for sequences as an additive extension of a single-letter metric. We provide a simple sufficient condition for Cd(W) = C(W), and more generally, for the equality of the d-capacities for two different decoding metrics d and d̄. This is followed by a sufficient condition for Ceo (W) = C(W). We then show that the lower bound on d-capacity given previously by past studies, is not tight in general but Cd(W) > iff this bound is positive. The 'product space' improvement of the lower bound is considered, and a 'product space characterization' of Ceo(W) is obtained. We also determine the e.o. capacity of a deterministic arbitrarily varying channel defined by a bipartite graph, and show that it equals capacity.

AB - We address the rate of transmission attainable on a given channel when the decoding rule is specified, perhaps suboptimally. We concentrate on decoders, termed d-decoders, which accept the codeword x 'closest' to the received sequence y in the sense of a metric d(x,y), defined for sequences as an additive extension of a single-letter metric. We provide a simple sufficient condition for Cd(W) = C(W), and more generally, for the equality of the d-capacities for two different decoding metrics d and d̄. This is followed by a sufficient condition for Ceo (W) = C(W). We then show that the lower bound on d-capacity given previously by past studies, is not tight in general but Cd(W) > iff this bound is positive. The 'product space' improvement of the lower bound is considered, and a 'product space characterization' of Ceo(W) is obtained. We also determine the e.o. capacity of a deterministic arbitrarily varying channel defined by a bipartite graph, and show that it equals capacity.

UR - http://www.scopus.com/inward/record.url?scp=0028720754&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0028720754&partnerID=8YFLogxK

M3 - Conference contribution

BT - IEEE International Symposium on Information Theory - Proceedings

PB - IEEE

ER -