Robust predictive quantization: A new analysis and optimization framework

Alyson K. Fletcher, Sundeep Rangan, Vivek K. Goyal, Kannan Ramchandran

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

Abstract

This work is focused on computing-via a deterministic optimization with linear matrix inequality (LMI) constraints, rather than a pseudorandom simulation-the performance of predictive quantization schemes under various scenarios for loss and degradation of encoded prediction error samples. The ability to make this computation then allows for the optimization of prediction filters with the aim of minimizing overall mean squared error (including the effects of losses) rather than to minimize the variance of the unquantized prediction error sequence. The main tools are recent characterizations of asymptotic state estimation error covariance and output estimation error variance in terms of LMIs. These characterizations apply to discrete-time jump linear systems in which the discrete portion of the system state is a Markov chain. Translating to the signal processing terminology, this means that the signal model is "piecewise ARMA," as is standard in many forms of speech processing.

Original languageEnglish (US)
Title of host publicationIEEE International Symposium on Information Theory - Proceedings
Pages427
Number of pages1
StatePublished - 2004
EventProceedings - 2004 IEEE International Symposium on Information Theory - Chicago, IL, United States
Duration: Jun 27 2004Jul 2 2004

Other

OtherProceedings - 2004 IEEE International Symposium on Information Theory
CountryUnited States
CityChicago, IL
Period6/27/047/2/04

Fingerprint

Speech processing
State estimation
Terminology
Linear matrix inequalities
Markov processes
Error analysis
Linear systems
Signal processing
Degradation

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Fletcher, A. K., Rangan, S., Goyal, V. K., & Ramchandran, K. (2004). Robust predictive quantization: A new analysis and optimization framework. In IEEE International Symposium on Information Theory - Proceedings (pp. 427)

Robust predictive quantization : A new analysis and optimization framework. / Fletcher, Alyson K.; Rangan, Sundeep; Goyal, Vivek K.; Ramchandran, Kannan.

IEEE International Symposium on Information Theory - Proceedings. 2004. p. 427.

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

Fletcher, AK, Rangan, S, Goyal, VK & Ramchandran, K 2004, Robust predictive quantization: A new analysis and optimization framework. in IEEE International Symposium on Information Theory - Proceedings. pp. 427, Proceedings - 2004 IEEE International Symposium on Information Theory, Chicago, IL, United States, 6/27/04.
Fletcher AK, Rangan S, Goyal VK, Ramchandran K. Robust predictive quantization: A new analysis and optimization framework. In IEEE International Symposium on Information Theory - Proceedings. 2004. p. 427
Fletcher, Alyson K. ; Rangan, Sundeep ; Goyal, Vivek K. ; Ramchandran, Kannan. / Robust predictive quantization : A new analysis and optimization framework. IEEE International Symposium on Information Theory - Proceedings. 2004. pp. 427
@inproceedings{2c40c2ab67d441aaa4e400d0e63ffaae,
title = "Robust predictive quantization: A new analysis and optimization framework",
abstract = "This work is focused on computing-via a deterministic optimization with linear matrix inequality (LMI) constraints, rather than a pseudorandom simulation-the performance of predictive quantization schemes under various scenarios for loss and degradation of encoded prediction error samples. The ability to make this computation then allows for the optimization of prediction filters with the aim of minimizing overall mean squared error (including the effects of losses) rather than to minimize the variance of the unquantized prediction error sequence. The main tools are recent characterizations of asymptotic state estimation error covariance and output estimation error variance in terms of LMIs. These characterizations apply to discrete-time jump linear systems in which the discrete portion of the system state is a Markov chain. Translating to the signal processing terminology, this means that the signal model is {"}piecewise ARMA,{"} as is standard in many forms of speech processing.",
author = "Fletcher, {Alyson K.} and Sundeep Rangan and Goyal, {Vivek K.} and Kannan Ramchandran",
year = "2004",
language = "English (US)",
pages = "427",
booktitle = "IEEE International Symposium on Information Theory - Proceedings",

}

TY - GEN

T1 - Robust predictive quantization

T2 - A new analysis and optimization framework

AU - Fletcher, Alyson K.

AU - Rangan, Sundeep

AU - Goyal, Vivek K.

AU - Ramchandran, Kannan

PY - 2004

Y1 - 2004

N2 - This work is focused on computing-via a deterministic optimization with linear matrix inequality (LMI) constraints, rather than a pseudorandom simulation-the performance of predictive quantization schemes under various scenarios for loss and degradation of encoded prediction error samples. The ability to make this computation then allows for the optimization of prediction filters with the aim of minimizing overall mean squared error (including the effects of losses) rather than to minimize the variance of the unquantized prediction error sequence. The main tools are recent characterizations of asymptotic state estimation error covariance and output estimation error variance in terms of LMIs. These characterizations apply to discrete-time jump linear systems in which the discrete portion of the system state is a Markov chain. Translating to the signal processing terminology, this means that the signal model is "piecewise ARMA," as is standard in many forms of speech processing.

AB - This work is focused on computing-via a deterministic optimization with linear matrix inequality (LMI) constraints, rather than a pseudorandom simulation-the performance of predictive quantization schemes under various scenarios for loss and degradation of encoded prediction error samples. The ability to make this computation then allows for the optimization of prediction filters with the aim of minimizing overall mean squared error (including the effects of losses) rather than to minimize the variance of the unquantized prediction error sequence. The main tools are recent characterizations of asymptotic state estimation error covariance and output estimation error variance in terms of LMIs. These characterizations apply to discrete-time jump linear systems in which the discrete portion of the system state is a Markov chain. Translating to the signal processing terminology, this means that the signal model is "piecewise ARMA," as is standard in many forms of speech processing.

UR - http://www.scopus.com/inward/record.url?scp=5044248654&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=5044248654&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:5044248654

SP - 427

BT - IEEE International Symposium on Information Theory - Proceedings

ER -