作者: Yahia Benmoussa , Jalil Boukhobza , Eric Senn , Djamel Benazzouz
DOI: 10.1109/DSD.2013.100
关键词:
摘要: Mobile devices such as smart-phones and tablets are becoming the most important channel for delivering end-user Internet traffic especially multimedia content. One of popular application is video streaming. The decoding process this compute-intensive responsible consumption a considerable part energy budget. Those mobile contain heterogeneous processing elements among-which we find Digital Signal Processors (DSP) General Purpose (GPP). In context, performance estimation those complex platforms difficult time consuming task when considering both hardware applicative parameters. paper, propose methodology developing unified high level analytical model embedded platforms. This based on experimental measurements conducted an low-power platform. developed describes H.264/AVC GPP DSP in terms bit-rate, clock frequency set comprehensive related coefficients. It achieves balance between too abstract detailed lower one while guaranteeing very good prediction properties (R-squared = 97%) tested videos. As use case, show that our allows to accurately determine bit-rate values which more energy-efficient than given