Title :
Universal source coding with a fidelity criterion
Author :
Kurtas, Erozan ; Salehi, Masoud
Author_Institution :
Dept. of Electr. & Comput. Eng., Northeastern Univ., Boston, MA, USA
fDate :
27 Jun-1 Jul 1994
Abstract :
Proposes a variable rate vector quantization algorithm which can adapt itself to variations in the source statistics by progressively updating the individual codewords. The proposed algorithm optimally parses the incoming source vectors and then quantizes them accordingly. After each quantization step, codeword updating is performed to track changes in the source statistics. No knowledge of the source statistics is assumed and no side information is transmitted
Keywords :
source coding; statistical analysis; tracking; variable rate codes; vector quantisation; change tracking; codeword updating; codewords; fidelity criterion; side information; source statistics; source vectors; universal source coding; variable rate vector quantization algorithm; Clustering algorithms; Probability; Source coding; Statistical analysis; Statistics; Sun; Testing; Vector quantization;
Conference_Titel :
Information Theory, 1994. Proceedings., 1994 IEEE International Symposium on
Conference_Location :
Trondheim
Print_ISBN :
0-7803-2015-8
DOI :
10.1109/ISIT.1994.395005
