TY - GEN
T1 - Optimal lossless compression
T2 - 2013 IEEE International Symposium on Information Theory, ISIT 2013
AU - Kontoyiannis, Ioannis
AU - Verdu, Sergio
PY - 2013/12/19
Y1 - 2013/12/19
N2 - This work1 deals with the fundamental limits of strictly-lossless variable-length compression of known sources without prefix constraints. The source dispersion characterizes the time-horizon over which it is necessary to code in order to approach the entropy rate within a pre-specified tolerance. We show that for a large class of sources, the dispersion of the source is equal to the varentropy rate, defined as the asymptotic per-symbol variance of the information random variables. We focus on ergodic Markov chains, whose optimal encodings are shown to be asymptotically normal and to satisfy an appropriate laws of the iterated logarithm.
AB - This work1 deals with the fundamental limits of strictly-lossless variable-length compression of known sources without prefix constraints. The source dispersion characterizes the time-horizon over which it is necessary to code in order to approach the entropy rate within a pre-specified tolerance. We show that for a large class of sources, the dispersion of the source is equal to the varentropy rate, defined as the asymptotic per-symbol variance of the information random variables. We focus on ergodic Markov chains, whose optimal encodings are shown to be asymptotically normal and to satisfy an appropriate laws of the iterated logarithm.
UR - http://www.scopus.com/inward/record.url?scp=84890403315&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84890403315&partnerID=8YFLogxK
U2 - 10.1109/ISIT.2013.6620525
DO - 10.1109/ISIT.2013.6620525
M3 - Conference contribution
AN - SCOPUS:84890403315
SN - 9781479904464
T3 - IEEE International Symposium on Information Theory - Proceedings
SP - 1739
EP - 1743
BT - 2013 IEEE International Symposium on Information Theory, ISIT 2013
Y2 - 7 July 2013 through 12 July 2013
ER -