Title :
Vector quantization of a parametric source
Author_Institution :
U.S. General Services Adm., USA
Abstract :
Interpreting the problem in terms of statistical risk, an approach is suggested in this paper for quantizing a parametric source with unknown statistics. Unlike universal quantizers, the suggested method achieves its optimal performance on a fixed dimension basis, and can be used to find a quantizer which minimizes the maximum quantization error when a source belongs to a class of parameterized sources
Keywords :
entropy; parameter estimation; rate distortion theory; statistical analysis; vector quantisation; maximum quantization error minimization; optimal performance; parametric source; rate distortion function; relative entropy; statistical risk; vector quantization; Distortion measurement; Entropy; Equations; Minimax techniques; Nearest neighbor searches; Parametric statistics; Probability density function; Rate-distortion; Robustness; Vector quantization;
Conference_Titel :
Communications, Computers and Signal Processing, 1997. 10 Years PACRIM 1987-1997 - Networking the Pacific Rim. 1997 IEEE Pacific Rim Conference on
Conference_Location :
Victoria, BC
Print_ISBN :
0-7803-3905-3
DOI :
10.1109/PACRIM.1997.620357
