Battery-Aware Multimedia Coding Optimization by Dynamic Frequency Scaling

Nowadays, multimedia application and video streaming have gained a great popularity because of the boasting development on mobile hardware, and battery driven devices have emerged with a tremendous speed. Due to the important issue of battery efficiency on mobile devices, many optimization algorithms have being proposed toward various of battery powered platforms and scenarios. Most of the provided solutions choose to aim at the minimization of energy usage under a given task scheduling by adjusting parameters reside in the processes of different schemes. However, the battery discharging characteristics and its instant output pattern are still ignored if the optimization is done only from high level adjustment. In this paper, we propose a battery-aware optimization framework toward the H.264 video coding by applying dynamic frequency scaling on hardware platform. The CPU frequency can be dynamically adjusted according to the instant status of battery in order to maximize the number of the coding frame. Experimental results indicate the efficiency and effectiveness of the proposed optimization framework.