Abstract: To deal with the high power consumption problem of mobile streaming media application, this paper proposes energy optimization method for mobile streaming media based on Markov decision processes for heterogeneous platform. By synthesizing the usages of different users, power consumption for video display, play time and current network state, CPU resources are reasonably distributed according to the principle of maximum award so as to reduce power consumption. The experimental result indicates that, compared with the system’s default scheduling strategy of heterogeneous multi-processing, the power consumption optimization method put forward in this paper can reduce the power consumption of streaming media video at the mobile terminal while maintaining its playback quality.

Key words: streaming media, low power consumption, Markov decision processes, heterogeneous multi-core architecture

摘要： 针对移动流媒体应用的高能耗问题，面向异构平台环境，提出了基于马尔科夫决策过程的移动流媒体能耗优化方法。通过综合不同用户使用习惯、视频播放功耗、播放时长及当前网络状态，基于最大奖励原则对CPU资源进行合理分配，以达到降低功耗的目标。实验结果表明，相比于系统默认异核多处理调度策略，提出的能耗优化方法能在保持流媒体视频播放质量的同时，有效降低其在移动终端的能耗。

关键词: 流媒体, 低功耗, 马尔科夫决策过程, 异构多核架构