Processing of high-quality streams is quite computationally expensive at the same time as video processing applications are more and more deployed in small embedded systems that traditionally exhibit limited processing and network resources, e.g. set top boxes, mobile phones or PDAs. Hence, for some combinations of streams and devices, the timing constraints on stream processing within a functional component will not automatically be met.
This paper analyses the functional components of a video stream processing and transmission with respect to their impact on the timing of end-to-end delivery chain processing. In particular it identifies parameters at a high level of abstraction, i.e., hiding the full internal details. The results can be used to make trade-offs between timeliness, resource usage and energy consumption at system design time.