Title :
Prediction and adaptive coding of sources with large or infinite alphabet
Author :
Ryabko, Boris ; Astola, Jaakko
Author_Institution :
Telecommun. & Comput. Sci., Siberian State Univ., Novosibirsk, Russia
fDate :
27 June-2 July 2004
Abstract :
This paper suggests a scheme of adaptive coding and prediction for a case where a source generates letters from an alphabet with unknown (and even infinite) size. This scheme can be applied along with any predictor; here we use the Laplace predictor as the main example. We consider a case of prediction for i.i.d. sources, but all results can be easily extended to Markov sources. The suggested scheme is described using the tree notation.
Keywords :
Markov processes; adaptive codes; prediction theory; probability; source coding; Laplace predictor; Markov source; adaptive coding; prediction scheme; tree notation; Adaptive coding; Adaptive signal processing; Application software; Computer science; Data compression; Information theory; Prediction methods; Probability; Redundancy; Statistics;
Conference_Titel :
Information Theory, 2004. ISIT 2004. Proceedings. International Symposium on
Print_ISBN :
0-7803-8280-3
DOI :
10.1109/ISIT.2004.1365063