Universal Anytime Coding

Stark C. Draper; Anant  Sahai

1st Internationl ICST Workshop on Control over Communication Channels

Research Article

Universal Anytime Coding

Cite: BibTeX Plain Text

@INPROCEEDINGS{10.1109/WIOPT.2007.4480120,
    author={Stark C. Draper and Anant  Sahai},
    title={Universal Anytime Coding},
    proceedings={1st Internationl ICST Workshop on Control over Communication Channels},
    publisher={IEEE},
    proceedings_a={CONCOM},
    year={2008},
    month={3},
    keywords={Communication channels  Communication system control  Computer errors  Context modeling  Delay  Error correction  Feedback  Maximum likelihood decoding  Sufficient conditions  Uncertainty},
    doi={10.1109/WIOPT.2007.4480120}
}

Stark C. Draper
Anant Sahai
Year: 2008
Universal Anytime Coding
CONCOM
IEEE
DOI: 10.1109/WIOPT.2007.4480120

Stark C. Draper¹^,*, Anant Sahai¹^,*

1: Dept. of Electrical Engineering and Computer Sciences University of California at Berkeley Berkeley CA 94720 USA

*Contact email: sdraper@eecs.berkeley.edu, sahai@eecs.berkeley.edu

Abstract

We show the existence of codes for anytime coding over compound channels - i.e. codes that operate without knowledge of the transition probabilities of the communication channel. The anytime error exponent achieved is the same as if the decoder were given the probabilities governing the realized channel and allowed to do maximum likelihood decoding. Because of the equivalence between system stabilization over noisy feedback channels and communication over noisy channels, the results immediately give sufficient conditions for stabilization over compound channels.

Keywords: Communication channels Communication system control Computer errors Context modeling Delay Error correction Feedback Maximum likelihood decoding Sufficient conditions Uncertainty

Published: 2008-03-31
Publisher: IEEE
Modified: 2011-07-31

: http://dx.doi.org/10.1109/WIOPT.2007.4480120

Universal Anytime Coding

Abstract

About EAI

Community

Publish with EAI