Although information theory is a comparatively young field of science (greatly influenced by Claude E. Shannon's paper of 1948), it has already undergone enormous development. Especially since the 80s of the last century, many techniques and methods have been explored as computers became available with increasing speed and power.
As side-effect the overview over all parts of data compression becomes more and more difficult. I hope my work will do its part to overcome this problem. I want to present a unified classification of compression techniques and methods. This systematisation intends to be a guidance for researchers and developers of new compression systems as well as an introduction for beginners in the field of data compression. It defines a clear structure, gives support for the understanding of compression and shows how to combine function blocks reasonably.
For a better understanding of the systematics we should at first clarify some terms. Usually, the words compression and coding are synonyms in the anglophone telecommunications community. Sometimes "data reduction" is used wrongly instead of "compression". I differentiate between these terms, because "data reduction" implies automatically that only the essential part of the data is kept and details are thrown away. "Compression" is in my classification a more global term whereas "coding" and "data reduction" mean only parts of compression as you will see in the sequel.