Design of Optimal Quantizers for Distributed Source Coding

  • Authors:
  • David Rebollo-Monedero;Rui Zhang;Bernd Girod

  • Affiliations:
  • -;-;-

  • Venue:
  • DCC '03 Proceedings of the Conference on Data Compression
  • Year:
  • 2003

Quantified Score

Hi-index 0.07

Visualization

Abstract

We address the problem of designing optimal quantizers for distributedsource coding. The generality of our formulation includes both the symmetricand asymmetric scenarios, together with a number of coding schemes, such asthe ideal coding achieving a rate equal to the joint conditional entropy of thequantized sources given the side information. We show the optimality conditions quantizers must satisfy, and generalize the Lloyd algorithm for its design.Experimental results are shown for the Gaussian scalar asymmetric case.