MPEG-2: a tutorial introduction to the systems layer

The jargon and the fundamental principles of the MPEG-2 systems layer are introduced and many optional syntax elements and many opportunities to include private (user-defined) syntax extensions enabling either multiplexing to be optimised to suit a particular application are introduced.

Abstract

The MPEG-2 systems specification describes how MPEG-compressed video and audio data streams may be multiplexed together to form a single data stream. This paper aims to introduce the jargon and the fundamental principles of the MPEG-2 systems layer. Two alternative multiplexing methods are specified for the MPEG-2 systems layer. The stream is biased towards the storage and replay of a single programme from a digital storage device while the transport stream is intended for the simultaneous delivery of a number of programmes over potentially error-prone channels. Both methods facilitate the inclusion of programme specific information detailing the programme(s) and elementary streams present. Both methods use a system of time stamps and clock references to ensure the synchronous replay of related elementary streams and correct buffer behaviour at a decoder. There are many optional syntax elements and many opportunities to include private (user-defined) syntax extensions enabling either multiplexing to be optimised to suit a particular application.