Title :
On strongly sequential compression of sources with abrupt changes in statistics
Author_Institution :
Dept. of Electr. & Comput. Eng., Utah Univ., Salt Lake City, UT, USA
Abstract :
An asymptotically optimal low-complexity strongly sequential compression scheme is proposed for universal lossless coding of memoryless sources with piecewise stationary abruptly changing statistics. The scheme is shown to achieve the lower bound for this universal coding problem even in a strongly sequential regime, where the horizon (i.e., the length of the data sequence to be encoded) is unknown when the algorithm starts to compress the data. Simulation results support the analytical results.
Keywords :
computational complexity; memoryless systems; sequential codes; source coding; statistics; abruptly changing statistics; asymptotically optimal sequential scheme; computational complexity; lossless coding; low-complexity; memoryless sources; piecewise stationary sources; source sequential compression; strongly sequential; Analytical models; Arithmetic; Cities and towns; Compression algorithms; Computational complexity; Gas insulated transmission lines; NASA; Statistics;
Conference_Titel :
Communications, 2003. ICC '03. IEEE International Conference on
Print_ISBN :
0-7803-7802-4
DOI :
10.1109/ICC.2003.1204571
