國立高雄大學圖資館 |

Language: English

Back

Universal discrete denoising under c...

Gemelos, George M.

Universal discrete denoising under channel uncertainty.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Universal discrete denoising under channel uncertainty.
Author:	Gemelos, George M.
Description:	107 p.
Notes:	Adviser: Thomas M. Cover.
Notes:	Source: Dissertation Abstracts International, Volume: 67-05, Section: B, page: 2738.
Contained By:	Dissertation Abstracts International67-05B.
Subject:	Engineering, Electronics and Electrical.
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3219278
ISBN:	9780542706790

Universal discrete denoising under channel uncertainty.
Gemelos, George M.

Universal discrete denoising under channel uncertainty. - 107 p.

Adviser: Thomas M. Cover.

Thesis (Ph.D.)--Stanford University, 2006.

The goal of a denoising algorithm is to recover a signal from its noise-corrupted observations. Perfect recovery is seldom possible; therefore we measure performance under a given fidelity criterion. For discrete signals corrupted by a known discrete memoryless channel, the Discrete Universal DEnoiser (DUDE) algorithm was recently shown to perform this task essentially optimally, without knowledge of the statistical properties of the source.

ISBN: 9780542706790Subjects--Topical Terms:

226981
Engineering, Electronics and Electrical.

Universal discrete denoising under channel uncertainty.
LDR:02093nmm _2200265 _450 001 180568
005 20080111103740.5
008 090528s2006 eng d
020 $a 9780542706790
035 $a 00311593
040 $a UMI $c UMI
100 0 $a Gemelos, George M. $3 264143
245 1 0 $a Universal discrete denoising under channel uncertainty.
300 $a 107 p.
500 $a Adviser: Thomas M. Cover.
500 $a Source: Dissertation Abstracts International, Volume: 67-05, Section: B, page: 2738.
502 $a Thesis (Ph.D.)--Stanford University, 2006.
520 # $a The goal of a denoising algorithm is to recover a signal from its noise-corrupted observations. Perfect recovery is seldom possible; therefore we measure performance under a given fidelity criterion. For discrete signals corrupted by a known discrete memoryless channel, the Discrete Universal DEnoiser (DUDE) algorithm was recently shown to perform this task essentially optimally, without knowledge of the statistical properties of the source.
520 # $a This thesis addresses the scenario where, in addition to the lack of knowledge of the source distribution, there is also uncertainty in the channel characteristics. We propose a discrete denoiser and establish its essential optimality under a worst case performance criterion, which we argue is appropriate for this setting. We also address the implementation and complexity of the proposed family of denoisers, establishing their practicality. In particular, we show that for many natural families of channels, the denoising rule can be mapped into a convex optimization problem, which can be solved efficiently. Finally, we present empirical results suggesting the potential of these schemes to do well in practice.
590 $a School code: 0212.
650 # 0 $a Engineering, Electronics and Electrical. $3 226981
690 $a 0544
710 0 # $a Stanford University. $3 212607
773 0 # $g 67-05B. $t Dissertation Abstracts International
790 $a 0212
790 1 0 $a Cover, Thomas M., $e advisor
791 $a Ph.D.
792 $a 2006
856 4 0 $u http://libsw.nuk.edu.tw:81/login?url=http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3219278 $z http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3219278