Adaptive Coding for Data Compression


Andrew Turpin
Department of Computer Science and Software Engineering, The University of Melbourne, Parkville 3052, Australia.

Alistair Moffat
Department of Computer Science and Software Engineering, The University of Melbourne, Parkville 3052, Australia.


Status

Proc. 22nd Australasian Computer Science Conference, Auckland, January 1999, 63-74.

Abstract

Many adaptive source coding techniques have been proposed to transmit messages drawn from a defined alphabet, each ideal for the particular application motivating its discovery. It is not obvious, however, which coding methods are appropriate for routine applications. In this paper we consider the practical tradeoffs that arise in choosing a coding method. A number of new and not-so-new techniques are briefly surveyed, and then results for compression performance, throughput, and memory consumption described, in each case with respect to a controlled experimental framework.