Conference
Minimum conditional entropy context quantization
Abstract
We consider the problem of finding the quantizer Q that quantizes the K-dimensional causal context C/sub i/=(X/sub i-t(i)/,X/sub i-t(2)/,...,X/sub i-t(K)/) of a source symbol X/sub i/ into one of M conditioning states such that the conditional entropy H(X/sub i/|Q(C/sub i/)) is minimized. The resulting minimum conditional entropy context quantizer can be used for sequential coding of the sequence X/sub 0/, X/sub 1/, X/sub 2/,.
Authors
Wu X; Chou PA; Xue X
Publisher
Institute of Electrical and Electronics Engineers (IEEE)
Publication Date
January 1, 2000
DOI
10.1109/isit.2000.866333
Name of conference
2000 IEEE International Symposium on Information Theory (Cat. No.00CH37060)