ABSTRACT
This chapter discusses the various approaches to compression, the tools that implement these, and the ways that tools are combined into complete systems. Advanced modulation techniques provide one part of the solution, but the most important enabling technology is video compression. Terminology varies, but most people differentiate between picture losses, such as a loss of sharpness, and artifacts that are visible effects in the output picture that were contributed by the compression system. Compression systems used for video are based on the assumption that the information to be coded represents real-world "photographic" images. The human visual system is very good at integrating over time, and it is generally possible to quantize B-frames more coarsely than reference frames without significant perceived impairment. The transform is an integer approximation to Discrete cosine transform, and spatial compression is aided by a sophisticated spatial predictor. The coding will use more bits than the calculated entropy of the data.
