Title :
Fuzzy estimation of unknown source model for universal coding
Author_Institution :
Inst. for Problems of Inf. Trans., Moscow, Russia
Abstract :
An “estimation-like” algorithm of sequential universal coding of sources with unknown model is proposed. It is shown that for the set of all context tree models, with restricted depths of contexts, the maximal individual redundancy of such coding decreases not slower than O(√((logn)/n)), n→∞, where n is the message length
Keywords :
fuzzy set theory; sequential estimation; source coding; trees (mathematics); context tree models; estimation-like algorithm; fuzzy estimation; maximal individual redundancy; message length; restricted context depths; sequential estimation; sequential universal coding; source coding; unknown source model; Arithmetic; Context modeling; Costs; Decoding; Upper bound;
Conference_Titel :
Information Theory Workshop, 1998
Conference_Location :
Killarney
Print_ISBN :
0-7803-4408-1
DOI :
10.1109/ITW.1998.706380
