Channel input adaptation via natural type selection

Sergey Tridenski, Ram Zamir

Research output: Contribution to journalArticlepeer-review

3 Scopus citations

Abstract

We propose an algorithm for computation of the optimal correct-decoding exponent, and its corresponding optimal input. The computation algorithm translates into a stochastic iterative algorithm for adaptation of the codebook distribution to an unknown discrete memoryless channel in the limit of a large block length. The adaptation scheme uses i.i.d. random block codes, and it relies on one bit of feedback per transmitted block. Throughout the adaptation process, the communication itself is assumed reliable at a constant rate R below the channel capacity C. In the end of the iterations, the resulting codebook distribution guarantees reliable communication for all rates below R + Δ, where 0 < Δ ≤ C-R is a predetermined reliability parameter affecting the speed of adaptation.

Original languageEnglish
Article number8840882
Pages (from-to)2078-2090
Number of pages13
JournalIEEE Transactions on Information Theory
Volume66
Issue number4
DOIs
StatePublished - Apr 2020
Externally publishedYes

Bibliographical note

Publisher Copyright:
© 1963-2012 IEEE.

Funding

Manuscript received November 4, 2018; revised August 24, 2019; accepted August 30, 2019. Date of publication September 17, 2019; date of current version March 17, 2020. The work of S. Tridenski and R. Zamir was supported in part by the Israel Science Foundation (ISF), under Grant 676/15 and in part by the US-Israel Binational and US-National Science Foundations (BSF-NSF) under Grant 2018690. This article was presented in part at ISIT2017 [1] and ISIT2018 [13].

FundersFunder number
BSF-NSF2018690
US-National Science Foundations
National Science Foundation1909423
Israel Science Foundation676/15

    Keywords

    • Arimoto algorithm
    • Blahut algorithm
    • Correct-decoding exponent
    • input distribution
    • unknown channels

    Fingerprint

    Dive into the research topics of 'Channel input adaptation via natural type selection'. Together they form a unique fingerprint.

    Cite this