|Supervisor:||Dr.-Ing. Christian Herglotz (Room 06.026)|
|Faculty:||Prof. Dr.-Ing. André Kaup|
Classical video codecs like MPEG-2 or H.264 aimed at keeping a high visual quality while at the same time spending as few bits as possible. State-of-the-art codecs like VP9 or HEVC additionally try to consider further criteria like complexity. The goal is not only to keep system requirements on a low level, but also to minimize energy consumption as common devices like tablets PCs or smartphones provide a limited battery capacity. A typical approach is to use processing time as an estimator for the energy consumption.
Measurements on a real system showed that there is a relation between time and energy consumption. Nevertheless it is unclear how this relation can be expressed. For example, it is well known that the energy consumption using memory accessing functions can increase dramatically while the processing time is barely affected.
The goal of this work is to find a valid relationship between energy consumption and processing time. As an approach, statistical observations as well as functional properties of a process may be investigated. The focus is en- and decoding of HEVC-coded videos. Measured data for energy consumption and processing time will be provided.