High-resolution distributed functional quantization

Vinith Misra, Vivek K. Goyal, Lav R. Varshney

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

Abstract

In traditional modes of lossy compression, attaining low distortion letter-by-letter on a vector of source letters X1N = (X1, X2, . . ..XN) ∈;ℝN is the implicit aim. We consider here instead the goal of estimating at the destination a function G(X1N) of the source data under the constraint that each Xi must be separately scalar quantized. The design of optimal fixed- and variable-rate scalar quantizers is considered under the assumptions of high-resolution quantization theory, yielding optimal point densities for regular quantizers. Additionally, we consider how performance scales with N for certain classes of functions. This demonstrates potentially large improvement from consideration of G in the quantizer design.

Original languageEnglish (US)
Title of host publication2008 Information Theory and Applications Workshop - Conference Proceedings, ITA
Pages531-534
Number of pages4
DOIs
StatePublished - Oct 6 2008
Externally publishedYes
Event2008 Information Theory and Applications Workshop - ITA - San Diego, CA, United States
Duration: Jan 27 2008Feb 1 2008

Publication series

Name2008 Information Theory and Applications Workshop - Conference Proceedings, ITA

Other

Other2008 Information Theory and Applications Workshop - ITA
CountryUnited States
CitySan Diego, CA
Period1/27/082/1/08

ASJC Scopus subject areas

  • Computer Science Applications
  • Information Systems

Cite this

Misra, V., Goyal, V. K., & Varshney, L. R. (2008). High-resolution distributed functional quantization. In 2008 Information Theory and Applications Workshop - Conference Proceedings, ITA (pp. 531-534). [4601035] (2008 Information Theory and Applications Workshop - Conference Proceedings, ITA). https://doi.org/10.1109/ITA.2008.4601035